New research topics
Past and current research topic
1. A Cost-Effective Automatic Temperature Control System Using Hybrid Fuzzy-PID Controller
Abstract — This paper presents an intelligent temperature control system that provides a fully automatic inverter air conditioning operation for a stable room temperature using hybrid fuzzy-PID controller. The system employs temperature sensors model DHT11 and Arduino microcontroller in order to obtain the temperature over awide area. The controllerprovides a control signal to VSDmodel VF-nC3 to control speed of the compressor motor.The experimental results show a fast and stable setting ofany specifically set temperatures comparing to theconventional commercial products. The proposed temperature control system offers a high reliability, stable temperature that applicable to the industrial plant.
2. A Smart Cost-Effective Real-Time Outdoor Car Parking Space Detection using Video Processing
Abstract — This paper presents a smart available parking lots detection system based on image processing technique. The system takes snapshot of the car park and using histogram normalization and blob analysis technique to identify the available parking lots. The outcome will be shown by LEDs in real time. Ten of green and red LEDs will were arrange as the same position as the parking lots to show the validity of the parking lots. The proposed system has both develop in software and hardware platform.
3. A Cost-Effective Radiopharmaceutical Dispenser with Real-Time Video Processing
Abstract — this paper presents the development of a cost-effective radiopharmaceutical dispenser with real-time video processing. Typically, a fully-automated radiopharmaceutical dispenser is expensive while a manual dispenser is harmful to operators due to radiation exposures. In the proposed system, the video camera sends the real-time video signals of the Fludeoxyglucos ([18F]-FDG) volume in a 5-ml syringe to a computer and the volume detection is subsequently performed through video processing using MATLAB Simulink. Air is slowly pushed through an extension tube from a 20-ml syringe to the [18F]-FDG vial until the 5-ml syringe is moved to reach the total [18F]-FDG required volume. The computer subsequently displays high-quality image for the operator that help a better vision of an amount of [18F]-FDG and therefore a one-time operation is sufficient. The proposed system offers a potential alternative to high-cost commercial radiopharmaceutical dispenser achieving a high precision and reducing operator’s radiation exposure.
4. An Efficient MPPT Tracking Method using Optimimal RBF Neural Network for a DC-DC Boos Converter Integrated with Photovoltaic Arrays
Abstract — This paper presents the Maximum Power Point Tracking (MPPT) method using optimal Radial Basis Function (RBF) neural network for a DC-DC boost converter integrated with photovoltaic arrays. The RBF neural networks employs Gaussian Functions in a hidden layer, and offers highly precised approximation of nonlinear continuous functions, particularly a solar cell systems that contains nonlinear voltage and currents based on the dynamically changed irradiation and temperatures. Optimizations are studied in terms of appropriate Gaussian functions, number of hidden layers, and learning characteristics. Simulations have been performed in a MABLAB Simulink through the use of a commercially available PV array model, i.e. a SunPower SPR-305-WHT model. The MPPT tracking are performed under irradiation variations between 50 W/m2 to 1,000W/m2 at and temperatures of 25oC. The RBF neural network enhanced control precision and response time compared to the perturbation-and-observation method, the incremental conductance method, and the multilayer perception networks.
5. The Implementation of Cost-Effective Data Acquisition System for Acoustic Emission Sensor using Variable Gain Preamplifier and STM32F4 Microcontroller Interface
Abstract — this paper presents the Data Acquisition (DAQ) system that particularly collects Acoustic Emission (AE) signals in the range of 1 kHz to 500 kHz from the sensor, and transmits to a personal computer for further signal analyses. The proposed system comprises two major components, i.e. a variable gain preamplifier and a STM32F4 microcontroller. The variable gain preamplifier is implemented using BiMOS operational amplifier model CA3140 which provides relatively high input impedance and high-speed performance. The signal is subsequently digitized by STM32F4 Microcontroller and transmits to data analysis software via serial communication (RS232). Simulations are performed through Orcad PSpice and a MATLAB Simulink. Signal quality analysis involves frequency response, Total Harmonic Distortion (THD), and Power Spectral Density (PSD). Experiments show that the proposed system offers a potential alternative to commercially available data acquisition systems.