Yul Y. Nazaruddin

Design and Implementation of Paired Pectoral Fins Locomotion of Labriform Fish Applied to a Fish Robot

In present, there are increasing interests in the research on mechanical and control system of underwater vehicles. These ongoing research efforts are motivated by more pervasive applications of such vehicles including seabed oil and gas explorations, scientific deep ocean surveys, military purposes, ecological and water environmental studies, and also entertainments. However, the performance of underwater vehicles with screw type propellers is not prospective in terms of its efficiency and maneuverability. The main weaknesses of this kind of propellers are the production of vortices and sudden generation of thrust forces which make the control of the position and motion difficult.On the other hand, fishes and other aquatic animals are efficient swimmers, posses high maneuverability, are able to follow trajectories, can efficiently stabilize themselves in currents and surges, create less wakes than currently used underwater vehicle, and also have a noiseless propulsion. The fish’s locomotion mechanism is mainly controlled by its caudal fin and paired pectoral fins. They are classified into Body and/or Caudal Fin (BCF) and Median and/or paired Pectoral Fins (MPF). The study of highly efficient swimming mechanisms of fish can inspire a better underwater vehicles thruster design and its mechanism.There are few studies on underwater vehicles or fish robots using paired pectoral fins as thruster. The work presented in this paper represents a contribution in this area covering study, design and implementation of locomotion mechanisms of paired pectoral fins in a fish robot. The performance and viability of the biomimetic method for underwater vehicles are highlighted through in-water experiment of a robotic fish.

Estimation of rice milling degree using image processing and Adaptive Network Based Fuzzy Inference System (ANFIS)

This paper describe a development of rice milling degree measurement system based on color analysis of rice sample. Rice Milling Degree is usually defined as the extent to which the bran layers of rice have been removed during the milling process. In Indonesia, rice quality is measured based on National Standard of Indonesia (SNI) of Milled Rice. Rice quality contains 11 variables resulting 5 categorizations of milled rice quality. Determination of rice quality is conducted manually by experienced inspector. This method has limitation in accuracy, objectivity and longtime measurement. This paper presents a measurement system of rice milling degree using image processing. Variety of IR-64 rice which 0%, 50%, 85%, 95% and 100% milling degree is used as sample and its image is taken using flatbed scanner. An RGB analysis then implemented to the sample and showed that the value of RGB is correlated with the value of rice milling degree in the sample. Adaptive Network Based Fuzzy Inference System (ANFIS) model then constructed using RGB value and normalized RGB value for better performance. Validation of ANFIS model using normalized RGB value resulting average error 3,55%.

An Experimental Investigation into the Design of a Robust Semi-Active Suspension System for a Quarter-Car Model

In this paper the experimental test of a robust semi-active suspension system using a quarter-car model has been investigated. The performance of the developed suspension system was evaluated under a sinusoidal road disturbance with amplitude of 2 mm (peak to peak) and 3 mm (peak to peak), and within the frequency test range of 1 – 9 Hz. The experimental results show that in the frequency test range of 1 – 4 Hz, the sprung mass acceleration of the semi-active suspension is smaller than the passive suspension. However, the sprung mass acceleration of the semi-active suspension is higher than the passive suspension, in the frequency test range of 5 - 9 Hz. Moreover, the unsprung mass acceleration of the semi-active suspension is smaller than its passive counterpart in all of the frequency test range 1 – 9 Hz.

Stabilization of Rotary Double Inverted Pendulum using robust gain-scheduling control

Development and application of linear parameter varying (LPV) controller for stabilization of upright position on rotary double inverted pendulum (RDIP) is presented. A robust gain scheduling approach is used for LPV controller synthesis. The stabilization problem of RDIP should handle three parameter uncertainties and one external disturbance. Lime-response evaluation shows that LPV controller outperform the LQR especially in terms of control performance and RMSE analysis.

Optimal control design with preview for semi-active suspension on a half-vehicle model

A design of optimal semi-active suspension control system applied to half-vehicle model is presented in this paper. The model has 4-degree of freedom (DOF) without involving the passenger seat. The optimal control strategy with constraints is used in designing the control, which involves six parameters representing comfort and safety. The obtained control signal has to be limited by constraints from the actuator (variable damper). The formulation of the problem and its solution are given as such that, in general, it is applicable to any vehicle system. Simulation results for seven chosen frequencies at vehicle speed of 60 km/h, illustrate the applicability of the proposed control strategy. Moreover, the results were also compared with the passive suspension system with regard to the comfort and safety criteria. It is found that the designed system shows a better safety factor than the passive suspension system.

Improving the performance of industrial boiler using artificial neural network modeling and advanced combustion control

In heat generation process, performance improvement is a critical factor and essential. An alternative solution is by designing an advanced combustion controller based on neural-predictive control strategy. However, for accomplishing such goal it requires adequate boiler model as well as combustion model. Although heat transfer and combustion processes in boiler are too complex to be analytically described with mathematical model, it can be approximated by artificial neural network model. This paper presents an alternative strategy to model the boiler and combustion process as well as proposes an advanced control strategy that takes the advantage of artificial neural networkpsilas ability as a universal function approximation. A feedforward neural network algorithm is applied to construct the models and the gradient descent technique seeks the optimal network weights, from which the nonlinear predictive control law under the reduced excess air level is derived. Direct application of this control strategy to real-time data taken from a running boiler system at an oil refinery plant demonstrated the benefit of the algorithm to improve the boiler combustion performance.

Improving performance of PID controller using artificial neural network for disturbance rejection of high pressure steam temperature control in industrial boiler

The temperature of high-pressure steam is very important to be controlled in order to perform other processes safely, especially for boiler-turbine system. Typically, PID regulator with fixed parameter is used for that purpose. However this method may usually deteriorate the control performance, particularly if the systems exhibit highly coupled behaviour. This paper will present the integration of intelligent control technique, especially artificial neural network, to challenge some deficiencies of PID regulator in dealing with such problem. The proposed control algorithm consists of a neural network controller, which is implemented parallel to the PID controller. The presented neural network controller involves HP steam temperature and its set-point as input and error control signal as a learning signal to be minimized. The ability of proposed algorithm is tested through step-like load disturbance into boiler plant model. Remarkable results have been obtained during this disturbance test. These results showed better performance to reject the disturbances compare with the controller which involves PID regulator alone.

Using Echo Ultrasound from Schooling Fish to Detect and Classify Fish Types

Fish finders have already been widely available in the fishing market for a number of years. However, the sizes of these fish finders are too big and their prices are expensive to suit for the research of robotic fish or mini-submarine. The goal of this research is to propose a low-cost fish detector and classifier which suits for underwater robot or submarine as a proximity sensor. With some pre-condition in hardware and algorithms, the experimental results show that the proposed design has good performance, with a detection rate of 100 % and a classification rate of 94 %. Both the existing type of fish and the group behavior can be revealed by statistical interpretations such as hovering passion and sparse swimming mode.

Wavenet Based Modeling of Vehicle Suspension System

An alternative modeling technique of vehicle suspension system which is based on an integration between wavelet theory and artificial neural network, or wavelet network (wavenet) is presented. Wavenet is a single hidden layer feedforward neural network, which uses wavelet basis function as an activation function. Polynomial windowed with Gaussian (POLYWOG) will be applied as the basis function. Wavenet parameters, such as weight, dilation and translation of the wavelet function will be optimized during its learning process, which is performed by a backpropagation algorithm. For minimizing the mean square error between model outputs and its observation, an iterative minimization method of gradient steepest descent is applied. Experimental evaluation of the proposed technique has been conducted using an input-output data collected from a running test vehicle. Observations by comparing the model responses with the actual output measurements revealed that satisfactory model matching were obtained which means that the models have captured the real basic features of the vehicle suspension dynamic characteristics

Applying neuro-fuzzy approach for ratio control of primary reformer in a petrochemical plant

This paper is concerned with a development of an alternative ratio control method using adaptive neuro-fuzzy approach and its implementation to control steam-gas ratio of primary reformer in a petrochemical plant in Gresik, Indonesia. Ratio control is used to ensure that two flows, i.e steam and gas, are kept at the same ratio even if the flows are changed. Adaptive neuro-fuzzy approach is implemented to model the dynamic inverse of the non-linear control valve system which controls the gas input to the reformer. The modeling is carried out in the learning phase using off-line technique, while in the design process of the neuro-fuzzy controller, an adaptive network will be employed as a building block. An hybrid learning rule is also used to minimize the difference between the actual and given desired trajectory. Results of simulation study demonstrate how the designed control technique performs well in tracking the changing of desired set-point. Performance comparison is also made between the designed and PI controllers.