In this work, the production of biodiesel via reactive distillation process has been modelled and simulated with the aid of ChemCAD for both steady state and dynamics. Also, the control of the process has been carried out using MATLAB/Simulink. In order to achieve this aim, dynamics data showing the response of biodiesel mole fraction in the column bottoms (controlled variable) to a change in reboiler duty (manipulated variable) and reflux ratio (selected disturbance variable) were extracted from the ChemCAD dynamic simulation of the developed process model and used to obtain the first-order-plus-dead-time transfer function relation between biodiesel mole fraction in the column bottoms, reboiler duty and reflux ratio with the aid of MATLAB. The open loop simulation was done by applying steps to the input variables (reboiler duty and reflux ratio). Furthermore, the set-point tracking and disturbance rejection control of the system were carried out using a PID controller tuned with Zeigler- Nichols, Cohen-Coon and trial-and-error techniques. It was observed that the controller parameters obtained by Zeigler-Nichols and Cohen-Coon tuning were not able to achieve the set-point tracking control of the system, and this necessitated the use of trial-and-error technique that was used to obtain the controller parameters used to handle the system in the desired manner for set-point tracking of maintaining the mole fraction of biodiesel at 0.9. Nonetheless, Zeigler-Nichols and Cohen-Coon tuning techniques were sufficient to successfully tune the process controller to carry out the disturbance rejection of the process. However, it was observed that the performance of Cohen-Coon tuning technique was better than that of Zeigler-Nichols tuning technique in the disturbance rejection control simulation as it had lower Integral Square Error and lower Integral Absolute Error values. It has, thus, been discovered that biodiesel could be produced in high purity via reactive distillation process, and the system could be efficiently handled to behave as desired using PID control system.
Keywords: Biodiesel, reactive distillation, PID control, set-point tracking, disturbancerejection, MATLAB, ChemCAD
1.1 Background of Study
Due to an increased demand of energy by the world population and the non-renewability of crude oil, the development of renewable energy generation techniques for future generations has gained great importance over the century (Madhu et al., 2012). One of these renewable energy has been identified to be biodiesel.
Biodiesel is a renewable, non-toxic, biodegradable substitute for diesel produced from crude oil. Generally, it is produced by transesterification of vegetable oils and animal fat by short chained aliphatic alcohols. Commercially, the production of biodiesel from vegetable oils and fats still have various drawbacks. Both batch and continuous processes utilize almost 100% excess alcohol than the stoichiometric molar requirement (3:1) in order to drive the transesterification reaction to completion and produce the maximum amount of biodiesel per unit consumption of oil (Kiss et al., 2008). At the end of the process, unreacted alcohol must be recovered by a separate distillation column. The use of a separate distillation column for alcohol recovery increases capital as well as operating cost. Therefore, there is the need to develop alternative means for the commercial production of biodiesel which minimizes cost without reducing the yield and quality of biodiesel produced. Reactive distillation is one of such alternative means.
Reactive distillation combines separation and reaction into a single vessel to minimize operation and equipment costs (Kiss et al., 2008). In this process, the products formed are removed as soon as they are formed. This characteristic makes it possible to overcome the
equilibrium thermodynamics of a reaction, reaching high conversion and selectivity. Thus, 1
it is particularly effective for reversible reactions such as the transesterification of vegetable oil and fats to biodiesel (He et al, 2006). However, the combination of reaction and separation into a single unit that resulted in many complexities of the process has made its dynamics and control study of this process a challenge to Process Engineers.
Dynamics in chemical engineering is the study of how process variables vary with time. As all real-life process variables vary with time, it is therefore important to study the dynamics of the biodiesel production process. Control is the external intervention needed to guarantee the satisfaction of operational requirements such as safety, production specifications, environmental regulations, operational constraints, economics (Stephanopoulos, 1984). Since the structure of biodiesel reactive distillation process is complex, due the need to maximize mass and energy raw materials, there is therefore need to develop a suitable control system for the process.
This research project is aimed at providing an outlook at the dynamics of biodiesel production by reactive distillation and developing a control system for the process by means of CHEMCAD and MATLAB modelling and simulation.
1.2 Research Problem Statement
Biodiesel is a valuable renewable fuel that can supplement and replace petroleum diesel in diesel engines. However, its cost of production by the reversible transesterification of vegetable oil and fats with alcohol by conventional means to achieve high purity of the product is relatively high. This high cost is a big problem that needs to be solved through provision of an alternative, novel, route and development of a reliable control method to make the process behave efficiently.
1.3 Aim and Objectives
The aim of this project is to model, simulate and control a reactive distillation process used for the production of biodiesel from the transesterification reaction between triolein and methanol. This aim would be achieved by realizing the following objectives:
• developing and simulating the ChemCAD steady-state model of the process,
• converting the steady-state model into a dynamic type to generate dynamic data,
• using the generated dynamic data to develop the process transfer functions with the aid of MATLAB,
• using the transfer function model of the process to obtain the tuning parameters of a PID controller,
• applying the PID controller to make the mole fraction of the biodiesel be at the desired set-point value.
This work is limited to employing ChemCAD and MATLAB to develop a model, simulate the model and carry out the open-loop and closed-loop simulations of the model for a reactive distillation used for biodiesel production from the transesterification reaction between triolein and methanol.
1.5 Justification of Research
The successful accomplishment of this work will enlighten the process engineers on the methods that can be utilized in handling a reactive distillation process very well to make it behave as it is desired at any time.