Fig 1 - uploaded by Andrei Torgashov
Content may be subject to copyright.
Source publication
Development, implementation, and operation of an advanced control system for a crude distillation unit are described. The system is based on Honeywell Profit® Controller software. The paper outlines the CDU/VDU-4 plant of Gasprom Neftekhim Salavat JSC (Bashkortostan Republic, Russia) as a control object, discusses project implementation phases, men...
Context in source publication
Context 1
... with attractive payback. The unit desalts and processes up to 4 million tonnes of stable gas condensate; its key products are unstable naphtha, kerosene, Diesel, vacuum gas oil, and long residuum. A mix of stable condensate and desalted crude oil can be also processed unless the mix's maximum density exceeds 0.805 g/cm 3 . The unit comprises of ( Fig. 1) a desalter section (electric dehydrators), crude and desalted crude preheat trains, evaporator E-200, pre-flash column K-210, atmospheric heater P-210, main atmospheric column K-220 with two strippings and five pumparound flows, vacuum heater P-310, and vacuum distillation column K-310 with two pumparounds. ...
Citations
... In their research, Fayruzov et al. developed a control system specifically for a crude distillation unit. The system has proven effective in ensuring optimal efficiency and safety, highlighting the importance of continued research in industrial control systems [55]. ...
The primary function of a crude distillation unit (CDU) within a petroleum refinery is to effectively segregate crude oil into its constituent fractions or products based on their respective boiling points. The Crude Distillation column often serves as the primary processing unit within most refineries, pivotal in producing a wide range of refinery products. This study examines research articles published between 2013 and 2023 that specifically investigate issues related to crude distillation units. The research endeavours to produce innovative designs and construct mathematical models to enhance production efficiency within this context. The research primarily centres on developing a mathematical model that accurately characterizes the distillation tower. This is achieved using either an Artificial Neural Network or a nonlinear model predictive control approach. The primary objective of simulation and optimization research is identifying optimal operating conditions, typically employing software tools such as Aspen HYSYS or PRO II. The corrosion treatment outcomes conducted at the tower's upper section were satisfactory. The study focused on the issue of corrosion in the overhead lines and pumps around exchangers. This design research aims to investigate potential modifications to the distillation tower's design or preflash process to optimize production outcomes.
... In this research the parameters expected to be monitored are only the feed flow rate and its temperature in the column. The research work mentioned in [14] described a case study on the implementation of advanced process control for crude distillation unit. In such study the research concern was on the way to control the distillation system process by the techniques of site survey, draft design, acceptance test and trial operation. ...
One of the key element of many industrial process plants is a crude oil distillation system. This system requires heat for the vaporization of a mixture of feed to the distillation column in vapor form. In the existing process of distillation, an abnormal change of heat is exhibited due to improper control and monitoring of disturbances affecting the plant. This would result for undesired loss of product and product purity to be reduced. The parameters that are expected to be considered in the analysis, model and control of a distillation system described under the proposed study are inlet feed temperature, distillation column temperature, feed composition, internal liquid and vapor composition, feed flow rate, reboiler temperature and an external reflux temperature to the distillation tower.
Controlling distillation column parameters with Adaptive model predictive control allows for the determination of the predicted future instant values of the plant outputs. Using such control technique, the controlled plant outputs such as the temperature of distillation column, feed preheater, re boiler and the upper reflux is properly controlled and their parameters are estimated using recursive least squares approach in the entire process adaptation mechanism. From the analysis and optimization work made on the proposed system, the efficiency of the plant outputs in tracking their corresponding set point has improved based on the value of the transient system parameters as well as the value of relative volatility of feed mixture to the column. As per the finding in the analysis of the process, 95.4% and 93.5% improvement on the set point tracking and to the amount of evaporation liquid feed has been obtained respectively. On the other hand an improvement on transient parameters has been achieved to all plant outputs. As per the result obtained from the analysis, the peak overshoot, settling time and peak time of the system response has found to be less than 40% including the effect of measured disturbance to the plant. Hence, entire process variable optimization has been performed using the parameters of the model predictive controller to provide the proper degree of stability. Finally the proposed method of study has compared with other control strategies through which the performance of the proposed design has been ensured.
... The estimation of product quality via soft sensors (Funatsu, 2018;Kim et al., 2013) is an inexpensive and attractive technique in industrial automation. Improving soft sensor evaluation methods may be the subject of any innovative APC or real-time optimization platform (Amrit et al., 2015;Fayruzov et al., 2017). ...
Traditionally, soft sensors are developed based on measurement data only, but here we consider an adaptive soft sensor that uses data generated from a fitted, first principles model of the distillation columns. The contribution of the paper is a procedure for moving window soft sensor design that incorporates a priori knowledge, which is especially suitable when the training sample is small and contains measurement errors. In addition, we propose a continuous adaptation of all model parameters based on new data, instead of the usual procedure of only updating the bias. The accuracy of the predicted product quality is investigated by calculating the coefficient of determination and root mean squared error for the test sample. Several approaches were considered, and we found that a constrained optimization approach was superior. The constraints on the model parameters of soft sensors are derived from a fitted, rigorous distillation unit model. The improved estimator quality resulted in the successful industrial application of advanced process control systems.
https://authors.elsevier.com/a/1ZleJx-vuKFXw
... They describe a technological process by connecting the measured technological parameters (temperature, pressure, consumption rate) to the results of laboratory analyses of an output variable and can be integrated into APC using a specialized software. Based on the assigned criteria, the optimal control action is formed in APC using a predictive model [3]. ...
... The reaction mixture is delivered to the SR beneath the lower catalyst layer, from where the outgoing vapors are directed to the distillation column C-1, on top of which distillate (D) is carried away, and the reaction mixture from the SR goes to the bottom distillation column C-2. To construct the model predicting the content of impurities in the output (bottom) product (BP), we used measured technological variables: u (1) is the reflux rate of C-1 (F 1 ); u (2) is the flow of reaction mass entering C-2 (F 2 ); u (3) is the pressure of the FR bottom (P 1 ); u (4) is the pressure of the SR bottom (P 2 ); u (5) is the temperature at the bottom of the column C-2 (T 1 ); u (6) is the temperature of vapors from C-2 (T 2 ); and u (7) is the temperature at the 5th stage of C-2 (T 3 ). ...
The problem of estimating the model parameters for predicting a quality indicator for an output product of a reactive distillation process is considered. The feedback loop in the prediction error of the output variable is used in the mathematical model. It is shown that this approach increases the accuracy of identifying a plant. The predictive model obtained is intended for being used as part of the improved system of the control of the technological process.
... The estimation of product quality via SSs (Ghadrdan et al., 2013) is an inexpensive and attractive technique in industrial automation. Improving SS evaluation methods may be the subject of any innovated APC or real-time optimization platform (Amrit et al., 2015;Fayruzov et al., 2017;Wang et al., 2017). In the present work, the sequence of industrial multicomponent distillation columns in the gas separation section of the fluidized catalytic cracking (FCC) unit at Gazpromneft-Omsk Refinery plant is investigated. ...
Traditionally, soft sensors are developed based on measurement data only, but here we consider a moving window (MW) soft sensor (SS) that uses data generated from a calibrated, rigorous model of the distillation columns of an FFC unit at Gazpromneft-Omsk Refinery. The contribution of the paper is that a procedure is developed for MW SS design that incorporates a priori knowledge, which is especially suitable when the training sample is small and contains measurement errors. In addition, we propose a continuous adaptation of all model parameters based on new data, instead of the usual procedure of only updating the bias. The accuracy of the predicted product quality is investigated by calculating the coefficient of determination (R 2) and root mean squared error (RMSE) for the test sample. Several approaches were considered, and we found that a constrained optimization approach was superior. The constraints on the model parameters of SSs are derived from a calibrated, rigorous distillation unit model. The improved estimator quality resulted in the successful industrial application of advanced process control (APC) systems.
The changes in the characteristics of the crude oil feed to an atmospheric distillation unit can influence the quality of the products. Also, the crude flow rate disturbances should be handled efficiently. An inferential control structure is adopted using RGA analysis where multiple tray temperatures are controlled. Then, a closed-loop stochastic optimization framework is proposed for tighter control and set of the decision variables is extended using the set-points of the decentralized PI controllers. The closed-loop simulations show improved disturbance rejection performance of the controllers against disturbances of feed flow rate and quality as well. Also, the challenges regarding the specifications of the product quality expressed as equality constraints are addressed. Both single and joint constraints cases are discussed.
An energy-efficient crude distillation unit (CDU) with a divided wall column was introduced to evaluate its performance compared to the conventional CDU. The large energy demand of the CDU in the U. S. - equivalent to more than a half of biofuel produced - was reduced by applying a divided wall column to the unit also known as the energy-efficient distillation column. The divided wall column lowers mixing at feed tray and raises the thermodynamic efficiency of the CDU. The performance evaluation of the proposed unit indicates that the unit saves 37% of heat supply over the conventional unit and cooling by 17%. The economic analysis shows a 9% of investment saving and a 26% decrease in the utility cost from the proposed unit. The thermodynamic efficiency of the proposed CDU is improved by 8%. The modification of the conventional CDU was minimal, suggesting an easy revamping of the current conventional CDUs.
