Artificial Neural Networks Controller for Crude Oil Distillation Column of Baiji Refinery

Abstract

A neural networks controller is developed and used to regulate the temperatures in a crude oil distillation unit. Two types of neural networks are used; neural networks predictive and nonlinear autoregressive moving average (NARMA-L2) controllers. The neural networks controller that is implemented in the neural network toolbox software uses a neural network model of a nonlinear plant to predict future plant performance. Artificial neural network in MATLAB simulator is used to model Baiji crude oil distillation unit based on data generated from aspen-HYSYS simulator. A comparison has been made between two methods to test the effectiveness and performance of the responses. The results show that a good improvement is achieved when the NARMA-L2 controller is used with maximum mean square error of 103.1 while the MSE of neural predictive is 182.7 respectively. Also shown priority of neural networks NARMA-L2 controller which gives less offset value and the temperature response reach the steady state value in less time with lower over-shoot compared with neural networks predictive controller.

Distributed Control Strategy with Smith's Predictor in a Pilot‐Scale Diabatic Distillation Unit

“…Aiming at efficient design and tuning distillation controllers, studies based on advanced techniques have been the focus of several works in the last decades. Karacan, Hapoglu, and Alpbaz [10] and Meng et al [11] used generalized predictive control; Karacan [12] applied nonlinear long-range predictive control; Rani, Singh, and Gupta [13] and Ahmed and Khalaf [14] implemented artificial intelligence through neural networks; and Miccio and Cosenza [15] did studies with fuzzy logic algorithms. All these studies were successful in reducing transient time; however, its reduction to zero is impossible, since the action of the reboiler must be propagated up to the last stage of the column, and the same occurs inversely with the action of reflux flow rate until the bottom of the column.…”