Optimization of Atmospheric Distillation Unit of Warri Refinery Using Artificial Neural Network and Exergy Rate Profiles

Funmilayo N. Osuolale, Ambrose N. Anozie

Abstract


In this paper, the operation of atmospheric distillation unit (ADU) of  Warri refinery was optimized using Artificial Neural Network (ANN) method of optimization and exergy rate profiles (ERP). Optimization of ADU exergy efficiency using nine operating variables and ANN method of optimization improved exergy efficiency from 33% to 53%. The vapour and liquid exergy rate profiles in the distillation column were used to reveal points of inefficiency within the column and as a retrofit tool to suggest possible column modification alternatives for energy efficient operations. The exergy rate profiles in the column were found to be crossing each other. Optimization of the ADU when the crossing of the exergy rate profiles in the column was removed further improved ADU exergy efficiency from 53% to 60%. Artificial neural network was shown to be a powerful and suitable optimization method for solving constrained optimization problems such as in atmospheric distillation unit with several operating variables with constraints. Exergy rate profiles depict the driving forces between the liquid and the vapour states in a column and were shown to be a suitable tool for further improvement of ADU exergy efficiency.


Full Text:

PDF

Refbacks

  • There are currently no refbacks.