Mohd Zakimi Zakaria

Comparison between multi-objective and single-objective optimization for the modeling of dynamic systems

Modeling input–output data representing a dynamic system is a challenging task when multiple objectives are involved. The developed model needs to be parsimonious yet still adequate. To achieve these goals, two objective functions, i.e. optimum structure and minimum predictive error, need to be satisfied. Most works in system identification only consider one objective function, i.e. minimum predictive error, and the model structure is obtained by trial and error. This paper attempts to establish the needs of a multi-objective optimization algorithm by comparing it with a single-objective optimization algorithm. In this study, two different types of optimization algorithms are used to model a discrete-time system. These are an elitist non-dominated sorting genetic algorithm for multi-objective optimization and a modified genetic algorithm for single-objective optimization. Simulated and real systems data are studied for comparison in terms of model predictive accuracy and model complexity. The results show the advantage of the multi-objective optimization algorithm compared with the single-objective optimization algorithm in developing an adequate and parsimonious model for a discrete-time system.

Effects of genetic algorithm parameters on multiobjective optimization algorithm applied to system identification problem

The growing interest in multiobjective optimization algorithms and system identification resulted in a huge research area. System identification is about developing a mathematical model for representing the system observed. This paper describes the effects of genetic algorithm parameters used in multiobjective optimization algorithm (MOO) that is applied to system identification problem. Two simulated linear systems with known model structure were considered for representing the system identification problem. The performance metrics used in this study are convergence and diversity metric. These metrics show the performance of MOO when GA parameters are varied. The simulation results show the effects of GA parameter on MOO performance. A right combination of GA parameters used in MOO is shown in this study.

Dynamic System Modeling of Flexible Beam System Using Multi-Objective Optimization Differential Evolution Algorithm

This paper proposes an algorithm called multi-objective optimization using differential evolution (MOODE) for providing the optimal mathematical model of flexible beam system. The main reason of developing a flexible beam system is to find an appropriate controller to control the vibration produced by this system. This dynamic system is treated as a black box where the acquired input-output data is used in the modeling processes. Two objective functions are considered for optimization; minimizing the number of term of a model structure and minimizing the mean square error between actual and predicted outputs. Nonlinear auto-regressive with exogenous input (NARX) model is used to represent the mathematical model of the investigated system. To obtain an optimal model for representing the dynamic behavior of flexible beam system, the model validity tests have been applied.

Design and development of a freezer and chiller delivery box

This paper presents an action research of designed and fabricated using well insulating materials in order to ensure the coolness inside the freezer and chiller delivery box is as good minimize temperature raised. The main purpose of this study is to develop freeze and chiller delivery box that will be able to keep fresh meat during travelling long delivery. A range of freeze and chill solutions exists for that must be kept within a specific temperature range throughout the supply-and-distribution chain. This will help to minimize the activeness of bacteria to spoil the meats, at the same time it can linger the duration for meats to spoil. All affecting parameter such as temperature inside the delivery box, heat transfer rate, and natural convection flow pattern has been studied to design and development of the delivery box. Finally, temperature distribution analysis has been conducted which showed that in operating condition inside temperature are suitable to keep the fresh meats condition.

Real-Time Self-Tuning Speed Controller: Performance Comparison between Engine Fuelled with Palm Methyl Esters and Petroleum Diesel

This paper investigates the performance of a real-time self-tuning speed controller designed to track and regulate at various engine speeds. The controller was tested with an automotive engine fuelled with petroleum diesel and and palm oil biodiesel (Palm Methyl Esters) within speed range of 1800 rpm to 2400 rpm. A self-tuning control algorithm based on pole assignment method together with on-line model parameters estimation strategy based on the recursive least squares method are adopted. The ability of the controller to track, regulate at various engine speed and also to reject disturbances applied for both type of fuel are compared and presented. The results confirmed that the controller performed very satisfactorily.

Evapotranspiration prediction using system identification and genetic algorithm

Reference evapotranspiration or ETO is important to provide information in planning and management of water resource system for irrigation purposes. Hence, its accurate estimation is of vital importance to assess water availability and requirements. This study explores the use of system identification approach and modified genetic algorithm (MGA) to model the evapotranspiration process under climatic data. The method is applied in modelling hourly evapotranspiration in central and southern region of Malaysia as a function of solar radiation, temperature, humidity and wind speed. The performance of the model is compared with the traditional Penman-Monteith (PM) method. Results from the study indicate that both the data driven is comparable with that of the PM method. The MGA models are dominated by temperature and solar radiation indicating that these two inputs can represent most of the variance. The results also show that the models are parsimonious and understandable, and are well suited to modelling the dynamics of the evapotranspiration process.