Project Title:

Distillation Column Flooding Predictor

Recipient:

George E. Dzyacky,
2ndpoint L.L.C.
524 Pinehurst Lane,
Schererville, IN 46375

Award Number:

DE-FC36-02ID14426

Subcontractors:

The University of Texas at Austin Center for Energy and Environmental Resources
Mail Code R7100
Austin, Texas 78713-7726
Dr. R. Bruce Eldridge
Phone: 512.471.7067

Contact:

George E. Dzyacky
2ndpoint L.L.C.
524 Pinehurst Lane
Schererville, IN 46375
Phone: 219.712.0434
Fax: 800.417.8940
ged@2ndpoint.com

Project Team:

DOE-HQ contact is Dickson Ozokwelu; Project Mentor is Dr. R. Bruce Eldridge, Center for Energy and Environmental Resources, The University of Texas at Austin. The DOE Project Officer is Bill Prymak, Golden Field Office, Golden Colorado.

Project Objective:

The goal of the project is to develop the Flooding Predictor, an advanced process control strategy, into a universally useable tool that will maximize the separation yield of a distillation column. This will be accomplished using a patented pattern recognition methodology the Flooding Predictor, to predict the onset of hydraulic flooding. The project is a multi-step approach that consists of industrial-scale validation, pilot plant-scale experimentation, dynamic model development, and pattern recognition model constant generation. Commercialization of the Flooding Predictor will be done by an industrial technology vendor with experience in supplying tools to the chemical industry.

Background:

The U. S. petroleum refining and chemical processing industries consume over 12 quadrillion BTUs of energy each year. Distillation is a low thermal efficiency unit operation (about 6% for easy separations) that currently accounts for 40% of the processing energy consumed in refining and continuous chemical processes. In spite of the high energy required for distillation, it is often chosen over other separation processes because of the relatively low initial capital investment, flexibility, and ability to yield high purity products. Currently, every barrel of crude oil is subjected to an initial separation by distillation, and nearly every chemical process requires distillation for product recovery or purification.

This high level of energy consumption and widespread utilization makes distillation column operations an extremely attractive area for optimization. The proposed research will develop a methodology that will optimize the energy input per barrel of feed to a distillation column. Widespread implementation of the technology will make a significant impact on the energy consumption of the chemical processing industry.

The research program will extend and validate an advanced process control strategy that utilizes a patented pattern recognition system to identify the onset of pre-flood conditions in distillation, absorption, and stripping columns. The strategy briefly relaxes column severity at the pre-flood state causing long-term operation to become significantly more stable and energy efficient. Potential energy wasting flood conditions are avoided, column stability is increased, and column throughput is increased.