Herbs have unique characteristics such as colour, texture and odour. In general, herb identification is
through organoleptic methods and is heavily dependent on botanists. It is becoming more difficult to
identify different herb species in the same family based only on their aroma . It is because of their similar
physical appearance and smell. Artificial technology, unlike humans, is thought to have the capacity to
identify different species with precision. An instrument used to identify aroma is the electronic nose.
It is used in many sector including agriculture. The electronic nose in this project was to identify the
odour of 12 species such as lauraceae, myrtaceae and zingiberaceae families. The output captured by the
electronic nose gas sensors were classified using two types of artificial intelligent techniques: Artificial
Neural Network (ANN) and Adaptive Neuro-Fuzzy Inference System (ANFIS). From the result, ANFIS
has 94.8% accuracy compared with ANN at 91.7%.
Special Protection Schemes (SPSs), are corrective action schemes that are designed to protect power
systems against severe contingency conditions. In planning of SPSs, protecting transmission network from
overloading issue due to critical situations has become a serious challenge which needs to be taken into
account. In this paper, a Special Protection and Control Scheme (SPCS) based on Differential Evolution
(DE) algorithm for optimal generation rescheduling has been applied to mitigate the transmission line
overloading in system contingency conditions. The N-1 contingency has been performed for different
single line outages under base and increased load in which generation rescheduling strategy has been
undertaken to overcome the overloading problem. Simulation results are presented for both pre-and
post system emergency situations. The IEEE 30-bus test system was utilised in order to validate the
effectiveness of the proposed method.
Space Vector Pulse Width Modulation (SVPWM) method is widely used as a modulation technique
to drive a three-phase inverter. It is an advanced computational intensive method used in pulse width modulation (PWM) algorithm for the three-phase voltage source inverter. Compared with the other PWM techniques, SVPWM is easier to implement, thus, it is the most preferred technique among others. Mathematical model for SVPWM was developed using MATLAB/ Simulink software. In this paper, the interface between MATLAB Simulink with the three-phase inverter by using Arduino Uno microcontroller is proposed. Arduino Uno generates the SVPWM signals for Permanent Magnet Synchronous Motor (PMSM) and is described in this paper. This work consists of software and hardware implementations. Simulation was done via Matlab/Simulink software to verify the effectiveness of the system and to measure the percentage of Total Harmonic Distortion (THD). The results show that SVPWM technique is able to drive the three-phase inverter with the Arduino UNO.
The grounding system of a lightning protection scheme is designed basically to avoid arcing and
dangerous step potentials. The grounding impedance of the system varies depending on soil structure and frequency. This paper describes the effect of harmonic impedance (also called frequency dependence of soil) on potential distribution under lightning strike to a metal tower with single grounding path, for different soil types. The results show that the peak value of ground potential rise (GPR) and step voltage (SP) may reach extremely hazardous values even at distances in the order of 90 m from the tower footing, especially when soil resistivity is high. Hence, we emphasise that, in contrast to power grounding, when designing of grounding systems that are meant to handle transient or high frequency currents as well, the frequency dependent soil parameters should be considered to avoid hazardous situations, especially at locations with a high probability of lightning strikes such as metal towers.
In this paper, cap and pin porcelain insulator was studied under an environment with different levels
of humidity. The electric field strength and voltage distribution profile along the insulator string was
simulated using a computational software package. In this study, ANSYS Maxwell based on the Finite
Element Method (FEM) was used to simulate the short standard insulator string. The short standard
insulator string was modelled as a five-unit cap and pin porcelain insulator that was stacked according
to the IEC 60383 standard. Different humidity levels measured using relative humidity is applied to
the insulator. From this simulation, the locations within the insulator under high electric field stress are
identified when different humidity is applied.
Visual inspection to locate metals embedded in walls or floors is impractical. Detection of these metals can only be done with a proper metal detection. Accordingly, the development of a magnetic imaging system based on giant magneto-resistance (GMR) sensors is presented for metal shape detection . This system is based on magnetic flux leakage testing (MFLT) principle for detecting the shape of ferromagnetic material specimens. The imaging system is constructed using 21 linear GMR sensors array as signals sensing unit (SSU). In this study, a few ferromagnetic SS400 mild steels specimens in various shapes are used as specimens. Image produced confirm system functionality in detecting and evaluating metal shapes.
Discrete Wavelet Transform (DWT)-based image compression techniques have been utilized in most of the earth observation (EO) satellites launched during the last few decades, since they have proved to be more efficient than other methods used previously with remote sensing multispectral imaging payloads. The efficiency of these techniques is mainly due to their high compression ratio that can be achieved while maintaining the quality of the compressed image. Also, they are considered multi-resolution compression techniques. However, these techniques are considered computationally demanding, due to their complex and sophisticated hardware. Due to the limited computational resources available on-board small satellites, they are considered one of the important criteria when choosing the satellite image compression method, along with the compression ratio and quality of the reconstructed image. Hence, an alternative DWT-based method was proposed, developed and implemented in this work with the aim of reducing the computational resources on-board a small satellite, replacing the regular DWT thresholding and quantization processes that are usually used to achieve lossy compression, with the zero-padding technique. This method will also help to control the change in the compression ratio and quality of the reconstructed image according to the end-user’s scientific needs of the satellite image. The
results of this work indicated, objectively and subjectively, that a decrease in the computational resources required on-board satellites was achieved by decreasing the processing time needed to complete the compression, without a significant difference in quality of the image reconstructed at the ground station.
In this paper, we proposed a new readout circuit in order to improve measurement output while reading the grouped resistive value in matrix array form. The purpose of having this circuit is to overcome the main problem in applying piezoresistive pressure sensor array for foot plantar application. This new approach is called Nodal Array Approach (NAA), a modified version of Wheatstone Bridge Circuit based on nodal analysis technique using Kirchcoff Current Law. The NAA calculates the sensors’ resistance values by solving simultaneous equations from reading voltages of the proposed readout circuit. Therefore, it is found that the readout circuit connection is of low complexity as it only uses resistive element as the major component of reading technique with only four iterations involved for each voltage nodes. Through simulation results, it shows that NAA is able to achieve high accuracy in obtaining a sensor’s resistance value by adhering to several limitations in order to avoid miscalculation (< 5% average calculation error).
The development of Transparent Solar Cells (TSC) has become the main focus of solar energy research in recent years. The TSC has a number of applications and make use of unexploited space such as skyscraper windows. In this paper, TSC is fabricated using commercially available titanium dioxide (TiO2) P25 to make a paste, which is deposited on FTO glass using screen printing and spin coating methods. The effects of the thickness of the TiO2 film on transparency are examined. The paste is synthesised in the Cleanroom and used in both methods of deposition. The final cell fabrication is a Dye sensitised solar cell (DSSC). The obtained transparency of the FTO glass is 83%, and after the deposition of TiO2 it is reduced to less than 80%. The overall transparency of the DSSC, which was made using the spin coating method, is 70% with an Isc of 9.5 mA and Voc 853mV.
This paper presents a transmission line (TL) modelling which is based upon vector fitting algorithm
and RLC passive filter design. Frequency Response Analysis (FRA) is utilised for behaviour prediction and fault diagnosis. The utilities of the measured FRA data points need to be enhanced with suitable modelling category to facilitate the modelling and analysis process. This research proposes a new method for modelling the transmission line based on a rational approximation function which can be extracted through the Vector Fitting (VF) method, based on the frequency response measured data points. A set of steps needs to be implemented to achieve this by setting up an extracted partial fraction approximation, which results from a least square RMS error via VF. Active and passive filter design circuits are used to construct the model of the Transmission line. The RLC design representation was implemented for modelling the system physically while MATLAB Simulink was used to verify the results.
Exoskeleton Robot is one of the most significant examples of human-oriented robotic devices.
Nevertheless, the main challenge remains the complexity of their mechanical design and human-robot interfaces. This paper is an outcome of a research to model and to simulate the support of mobility of an elderly people using exoskeleton. Exoskeleton is developed in order to complement the corporal deficiencies of an elderly person in standing up and sitting down. When the natural joint torques is integrated with the exoskeleton’s torque the result is in an overall torque that is comparable to that of a physically normal person. This work focuses on standing-up and sitting-down movements. Appropriate simulation models are formulated and their performances examined against measured data. The results with PID control show that at different speed of standing up and sitting down, the joint torques can be compromised. This is done within allowable limits.
Detection and quantification of DNA is critical to many areas of life sciences and health care, from
disease diagnosis to drug screening. The transduction of DNA through electrochemical methods have a fast response rate and with a conductometric device like the silicon nanowire which can be fabricated to have a similar diameter of the DNA molecule being targeted, detection is real-time. Critical to this is the interfacing of a current-source and an amplifier capable of achieving a maximum of 10 pico ampere input bias. In this project, we fabricated a silicon nanowire using the top down approach and built a circuit that can mimic the output signal as low as 12 nA and achieved a gain of 1 million to be interfaced with the nanowire for real-time DNA detection.
The core losses in a three phase transformer can be significantly reduced by improving the core joint geometry. The researchers were applied numerous types of T-joint designs in order to reach the optimum design that can be used in three phase transformer to reduction the losses. Two types of T-joint design are presented in this paper; T-joint with 90° butt-lap design and T-joint with 45° mitered design. A 3-phase distribution transformer was simulated in 3D using Ansys Maxwell software. The core loss for a three-leg three phase transformer rated 1000 KVA and the flux density distribution are investigated. The simulation results show the core losses were increased up to 3% and the flux density was increased to reach more than 22% flux density become higher when using T-joint with 90° butt-lap design as compared with T-joint with 45° mitered design.
This study presents the sensitivity of graphene nanoribbon (GNR) when exposed to ammonia gas at room temperature. Alumina were used as a substrate and coated with GNR as sensing film for ammonia gas detection. Four different concentration of GNR in the category of maximum, high, low, and minimum were prepared. Each category of GNR will be dispersed on alumina substrate with area of 1cm2 and 4cm2. 30nm of gold contacts are sputtered on both ends of the sensing film. The ammonia gas can be detected by measuring the changes in resistance. The GNR as ammonia sensor shows good responses at room temperature. In repeatability test, maximum GNR shows least variation when exposed to ammonia with the value of 1.01% (4cm2) and 2.12% (1cm2). In a sensitivity test, 0.25% to 1.00% of ammonia gas was used and tested on maximum GNR. Maximum GNR on 4cm2 substrate shows higher sensitivity as compared to 1cm2. Reaction time of GNR on ammonia gas decreased as the concentration of ammonia increased. Larger surface area of sensing element required lesser reaction time.
Preventive tests and diagnosis of in-service power transformer are important for early fault prediction and increased reliability of electricity supply. However, some existing diagnostic techniques require transformer outage before the measurement can be performed and need expert knowledge and experiences to interpret the measurement results. Other measurement techniques such as chemical analyses of insulating oil may cause significant variance to measurement results due to different practices in oil sampling, storage, handling and transportation of oil. A cost-effective measuring technique, which is simple, providing fast and an accurate measurement results, is therefore highly required. The extended application of Polarisation and Depolarisation (PDC) measurement for characterisation of different faults conditions in-service power transformer has been presented in this paper. Earlier studies on polarisation and depolarisation current of oil samples from in-service power transformer shows that depolarisation has provided significant information about the change of material properties due to faults in power transformer. In this paper, a new approach based on Depolarisation Current Ratio Index (DRI) was developed for identifying and classifying different transformer fault conditions. The DRI at time interval of 4s to 100s was analysed and the results show that DRI of depolarisation current between 5/100s and 10/100s provides higher correlation on the incipient faults in power transformer.
Tuberculosis (TB) is the second biggest killer disease after HIV. Therefore, early detection is vital to
prevent its outbreak. This paper looked at an automated TB bacteria counting using Image Processing technique and Matlab Graphical User Interface (GUI) for analysing the results. The image processing algorithms used in this project involved Image Acquisition, Image Pre-processing and Image Segmentation. In order to separate any overlap between the TB bacteria, Watershed Segmentation techniques was proposed and implemented. There are two techniques in Watershed Segmentation which is Watershed Distance Transform Segmentation and Marker Based Watershed Segmentation. Marker Based Watershed Segmentation had 81.08 % accuracy compared with Distance Transform with an accuracy of 59.06%. These accuracies were benchmarked with manual inspection. It was observed that Distance Transform Watershed Segmentation has disadvantages over segmentation and produce inaccurate results. Automatic counting of TB bacteria algorithms have also been proven to be less time consuming, contains less human error and consumes less man-power.
This paper discusses the harmonic analysis of the AC-to-DC uncontrolled converters commonly used in electric vehicles charging station. The aim of this paper is to model and simulate different rectifier models in addition to explaining the differences in input current harmonics, the total Harmonic Distortion (THD) as well as the power factor (pf). The converter configurations include single-phase bridge, 6-pulse and 12-pulse rectifier circuits. The single phase is normally used for electric scooter charging, while three-phase converters can be used for both electric bus and car charging. The circuit configurations of the rectifiers were modelled and simulated using Matlab R2014a to achieve the objective of the study. The results revealed that the THD levels were extremely high which is unacceptable if the system is connected to the utility grid.
Contaminated and ageing transmission line insulators often suffer from temporary or permanent loss of their insulating properties due to flashover resulting in power system failure. Surface discharges are precursors to flashover. To pre-empt any occurrence of flashovers, utility companies monitor the conditions of their insulators. There are numerous insulator surface monitoring techniques such as Leakage Current, Acoustics, and Infrared. However, these techniques may not be suitable for in-situ condition monitoring of the insulators as they are prone to noise, affected by environmental conditions or contact methods. Monitoring of the UV signals emitted by the surface discharges of these insulators has been reported to be a promising technique. However, comprehensive studies on this technique is lacking, especially on aged insulators. This paper investigated the UV signals of contaminated and aged insulators detected during surface discharge activities using UV pulse method. The time and frequency domain of the UV signals were analysed for a group of insulator samples having varying levels of contamination and phases of ageing. Results show that there is a strong correlation between the contamination level and ageing of the insulators with the amplitude and harmonic components of the UV signals. This correlation can useful to monitor in-service insulator surface conditions.
Digital predistortion is one of the most widely used techniques to linearize a power amplifier (PA) to
reduce the error vector magnitude (EVM) distortion and spectral regrowth. By far, the lookup table (LUT) predistorters are most frequently used scheme to mitigate the effects of non-linear power amplifier. In this paper, a new algorithm of joint-polynominal LUT predistorter which attains the best linearization performance is proposed. The algorithm employs the hermite interpolation LUT, which has a higher accuracy of interpolation. Simulation results show that the proposed method provides a better rejection of EVM distortion and an improvement of 30-40% of adjacent channel leakage ratio (ACLR) for the wideband code division multiple access at a minimal memory usage.
Multimode multiband connectivity has become a de-facto requirement for smartphones with 3G
WCDMA/4G LTE applications. In transceivers, multiband operation is achieved by selecting an output from two or more signal path targeting for a specific frequency range in parallel or by using switched capacitor/inductor. In this paper, design methodology of 280nm CMOS switch is presented. Design optimization of RF CMOS switch is presented which is deciding proper selection of CMOS transistor parameters and switch size as per external circuit parameters. The CMOS switch of a 5-transistor stack with W/L=1200μm/280nm provides insertion loss < 0.6dB and isolation loss >14dB. The switches designed when implemented in a multiband power amplifier (PA) exhibits 36dB gain at 1900MHz high-band and 34.5dB gain at 900MHz low-band with 27.5dBm peak power at both bands. The switch design methodologies presented in this paper should be of use in designing various blocks in emerging multiband transceiver applications.