Zakiah Mohd Yusoff

Modeling of steam distillation pot with ARX model

This paper present a autoregressive with exogenous input (ARX) was applied in modelling of steam distillation pot system. The system of steam distillation pot are design with special specification interm of material and shape additional with induction heating system that provide a new technologies in extraction of essential oil to fullfill demand for this industry. The performance of ARX was investigates and the training data that provide robust model and fresh data set was evaluate. ARX model was estimated by means of prediction error method and training function of Levenberg-Marquardt algorithm. Data was collected with probability band 0.2 of PRBS input signals. By ARX model structure, the input output data are model. The performance of the model based on coefficient of determination (R2), mean square error(MSE) and loss function of one-step-ahead and residuals. In overall, results shows that ARX have capability in capturing dynamic of the system.

MOdeling Of Steam Distillation System Using Hammerstein-Wiener model

This paper presents a new method to model a steam temperature in distillation system by using system identification. Three nonlinear models have been compared, i.e. a Hammerstein model, a Wiener model and a Hammerstein-Wiener model. In this work, we propose the utilizing of the piecewise-linear and sigmoid network Hammerstein-Wiener model for single-input single output processes. All the models have been optimized with respect to initial state, search criterion and number of iterations. The testing of the trained model will be based on percentage of best fit (R2), Final Prediction Error (FPE) and loss function (V). Among three model tested, the most accurate model is the Hammerstein-Wiener model with piecewise linear and sigmoid network estimators. This model produce highest percentage of best fit, the lowest FPE and loss function.

ARX modeling for down-flowing steam distillation system

This paper presents a logical continuation of our work by proposing the utilizing of ARX model by using pseudorandom binary sequence (PRBS) as the input signal and temperature as an output. The down-flowing steam distillation column developed in this work is intended for extracting the essential oil by steam distillation technique to improve the quality of output yield and get high quantity of essential oil as compare to the conventional steam distillation. The LabView 2010 was used to design the block diagram and front panel for controlling the data collection. The modeling is done by using model order 1 and model order 2. From this analysis the data collected show that the best fit from modeling using ARX model is greater than 98% for both model orders. From the result, ARX model is adequate to model the behavior of heating process is closed to the real system.

Steam temperature control of hydro-diffusion essential oil extraction system using hybrid-fuzzy plus PID controller

In extracting the essential oils, some of the influential factors need to be considered which can affect the yield and quality of the oils. Many studies have performed on several influential parameters and dictated that the temperature of extraction plays an important parameter that can give large effect on yield and quality. In order to regulate the temperature, a suitable controller is required. However, until now, very little publications stress on the control development of essential oil extraction system. This work proposed the simulation comparative studies on control performance of hybrid fuzzy plus PID controller using 3, 5 and 7 membership functions to regulate the steam temperature of hydro-diffusion essential oil extraction system. The ARX model was used to represent the system dynamic. The effectiveness of the proposed controller was tested via robustness test. Based on the simulated outcomes, the step responses analysis shows that the controllers performed very well. From the set point tracking test, the developed controllers were able to track the changes of set point whether in small or large set point change. There were also true for load disturbance tests.

Steam temperature control for steam distillation pot using model predictive control

This paper present a model predictive control (MPC) that was proposed to implement at plant steam distillation pot with induction heating system. By using the ARX model structure, the model predictive control was design to control of steam temperature. Since the steam temperature play important role during the extraction of essential oil, important to researcher to find the control technique to optimize the product yield and quality. The development and implementation of an ARX (Auto-Regressive exogenous) process model that will be used to design a Model Predictive Control (MPC) for steam distillation extraction system. MPC has been proposed as a controller to regulate the system that can preserve the optimum operation temperature besides minimizing the energy that is used to power up the plant. MPC tuning was examined with several values of prediction horizon with the similar default parameter to achieve the optimal setting for better controllers performance. Simulation results show that MPC is able to control the steam temperature in more efficient way using a first order ARX model.

Implementation of real -time steam temperature control for high efficient induction- based extraction essential oil using hybrid fuzzy-PD plus PID controller

Steam temperature is well known as one significant parameter in process of extraction essential oil that can contribute to increase the output yield and quality of the product. In this study, induction heating system was adopted in steam distillation pot to replace the conventional heating element. Induction heating system was preferred is due to high efficient, clean and save energy. In term of temperature control, this study implemented hybrid fuzzy PD plus PID at steam distillation pot with induction heating system for essential oil extraction system. Initially, the system was modeled to represent the system using ARX structure with PRBS was triggered to the system as an input. The parameter estimation and tuning is derived by simulation for HFPP control scheme. The performance of HFPP with 25 and 49 rules is compared by step response test. From experiment of real time control, demonstrate the proposed HFPP using 49 rules giving a better performance based on settling time, rise time, percent overshoot and RMSE value.

Characterization of down-flowing steam distillation system using step test analysis

This paper presents an application of step test in analysing the characterization and to observe the dynamics of new system which is down-flowing steam distillation system. Steam temperature data were collected in two condition: (a) without sample in the steam tray, (b) with sample in the steam tray-Cymbopogon Nardus (serai wangi). The down-flowing steam distillation column developed in this work is intended for extracting the essential oil by steam distillation technique to improve the quality of output yield and get high quantity of essential oil as compare to the conventional steam distillation. Step test is a conventional method that was conducted to observe the process dynamics of heating such as time delay, Td and dominating time constant, Tc. From this analysis, the dynamic of down-flowing steam distillation system for both situations are consistent along the process running.

Essential oil extraction with automated steam distillation: FMRLC for steam temperature regulation

Fuzzy model reference learning control (FMRLC), a model based controller, was implemented in steam distillation extraction oil process. The FMRLC was designed to regulate steam temperature inside the distillation column during extraction process. Maintaining the temperature to a certain degree for the period of extraction process is very important because the temperature will influence the output quality and quantity of essential oil. In this study, the FMRLC was simulated based on ARX modeled plant to regulate process steam temperature up to 85°C. The simulation was based on tuning method to obtained optimized FMRLC parameters. Then, the simulated result was validated by real process implementation.

Real time PID control for hydro-diffusion steam distillation essential oil extraction system using Gradient Descent tuning method

This paper present the implementation of real time PID controller for hydro-diffusion steam distillation essential oil extraction system based on comparison of Gradient Descent (GD) and Ziegler Nichols (ZN) tuning methods. The first order of Auto Regressive Exogenous (ARX) model was used to describe the behavior of the temperature system and will be use in the controller design. A PID controller is expected to execute a robust response towards parameters changes and better control of steam temperature during distillation process. The system had been evaluated based on rise time, % overshoot, settling time and root mean square error (RMSE). The temperature control was achieved by controlling the voltage fed to the heater ranging from 0V to 5V via digital-to-analogue converter (DAC). Robustness test of the PID controller are based on: i) introduce disturbance and ii) set-point tracking. From the result, the performance of PID controller using GD tuning method reveals that this controller can be adapt for the system because being sensitive to parameters changes and robust as the response can compensate the load disturbance and set point change.

The self-tuning fuzzy PID to regulate the steam temperature in real-time process

Conventional steam distillation is lack of control in order to satisfy the process requirement. In steam distillation process is regarded as the most significant parameter that contributes to the amount of output yield and quality of oil. An appropriate controller needs to integrate to the plant in order to regulate the steam temperature at desired or appropriate condition. This study demonstrated several controllers which are self-tuning fuzzy PID (STFPID) and hybrid fuzzy PD with PID (HFPPID) controller in order to improve the output controller performance compared to the conventional PID controller. Initially, this controller is proposed to reduce the overshoot of output, improve the rise and settling time and reduce a steady-state error of the step response. The performance from demonstrated controller shows that STFPID has the ability to improve process rise time, settling time and reduce of steady-state error and process overshoot compared to others.