Noraliza Hamzah

Internal fault classification using Artificial Neural Network

The main objective of this project is to create an intelligent model using image processing techniques in order to categorize the internal fault to four categories, which are low, intermediate, medium and high. Sample of internal fault location are captured using infrared thermography camera in which the RGB color image are stored and processed using Matlab. Processing involves impixelregion which includes creating a Pixel Region tool associated with the image displayed in the current figure, called the target image. This information is then being used to train a three layer Artificial Neural Network (ANN) using Levenberg Marquardt algorithm. A total of 168 samples are used as training whilst another 168 samples are used for testing. The optimized model is evaluated and validated through analysis of performance indicators frequently used in any classification model.

Prediction of water quality index (WQI) based on artificial neural network (ANN)

This paper investigates the effectiveness of artificial neural network models for predicting the water quality index for rivers in Malaysia. The network was trained with reference to seven major parameters for the determination of the water pollutant index, water quality index and water quality class, for Malaysian rivers in Pahang and Selangor. The data collected comprises of data for the previous three years, beginning from 1999. The water quality index plays an important role in evaluating the water quality of rivers. The artificial neural network simplifies and speeds up the computation of the water quality index, as compared to the currently existing method. By optimizing the calculation, a significant saving in terms of money and time can be achieved. Artificial neural network models with different learning approaches, such as back propagation neural network, modular neural network and radial basis function network, are considered and adopted to model the water quality index.

Transient stability analysis on Sarawak's Grid using Power System Simulator for Engineering (PSS/E)

Transient stability analysis has become one of the major analyses in the power system to ensure the system stability to withstand a major disturbance. The effect of transient occurrence can lead to malfunction of electronic control equipment. Transient analysis can be conducted using simulation software package. One of the commercial simulation software package that used by industry worldwide is Siemens Power System Simulation for Engineering (PSS/E). This research paper highlights the usage of PSS/E to analyze Sarawaks Grid System stability using the simplest dynamic model that has been embedded into the program. To observe transient analysis using PSS/E, basic machine model such as generator, exciter and governor were used by varying default data in the program to find the best simulation output. This paper also analyzed the theory of critical clearing time (CCT) of fault occurrence between a transmission line near to the generator and far from the generator. CCT appears to be lesser when the fault occurs at a transmission line near to the generator. On the contrary, when fault occurs far from the generator, the duration of CCT is greater. The stability of the system is observed based on the machine rotor angle, machine speed, output electrical power and terminal voltage.

Transient Stability Analysis of the IEEE 14-Bus Test System Using Dynamic Computation for Power Systems (DCPS)

Transient Stability Analysis (TSA) is a major analysis in the operation of power systems, due to the increasing stress on power system networks. One of the main goals of this analysis is to gather critical information, such as critical clearing time (tCCT) of the circuit breakers for faults in the system. tCCT is defined as the maximum time between the fault initiation and its clearing, such that the power system is transiently stable. This paper presents a transient stability analysis of the IEEE 14 bus test system using Dynamic Computation for Power Systems (DCPS) software package. This C++ based software package has the ability to handle systems up to 1000 buses and 250 generators, providing an alternative to expensive commercial software packages. To analyze the effect of the distance of the fault location and critical clearing time on the system stability, a three-phase fault has been applied at five different locations in the system. The stability of the system has been observed based on the simulation graphs of terminal voltage, machines rotor angle, machines speed and output electrical power. The simulation results showed that tCCT decreases as the fault location becomes closer to the main generator.

A study on THD reduction by active power filter applied using closed-loop current controlled AC-AC SPMC topology

An active power filter with single phase matrix converter topology is presented to reduce the THD of the input current, output current and output voltage of the system which is the main analysis in this paper. AC-AC converter topology switching strategy is used in the SPMC which is controlled by the current control closed-loop circuit or also known as active current wave-shaping technique. In order the AC-AC SPMC to operate and synthesize as an active power filter, the input supply current is being compensated by current control closed-loop circuit to produce the SPWM signal to be injected to the switches used which are the IGBTs. The work has been done using Matlab/Simulink simulation and the results are being compared to the same simulation done in open-loop method without using the active current wave-shaping technique.

Effect of impedances line length to voltage sag propagation

Occurrences of faults in power system contribute extensive voltage sags (dips) in a network. These occurrences have introduced problems to consumers mainly at 415V where sensitive control equipment is located. This paper investigates the propagation and travelling of voltage sag event under the influence of transformer connections and impedance line length. PSS/ADEPT (Power System Simulator Advanced Distribution Engineering Productivity Tool) was used to construct and simulate the test system for the study. Installation of monitoring devices cannot be done in every part of voltage level because it increases cost and is difficult to manage. The test system used allows major variations in network parameter for understanding its influence on voltage sag propagation. It is important to understand on how voltage sag propagates and travels through different voltage levels in power system network as whether it reaches and harms consumer equipment.

Classification of transient in power system using support vector machine

In this paper, application of SVM to classify disturbances in power quality is discussed. Power system transient can pose a serious threat to the reliability of power system apparatus and sensitive loads. There are numerous causes of power system transient namely short circuits, capacitor bank switching, switching of large inductive loads that include motors and transformers as well as lightning. Firstly, an IEEE 30 bus system is modeled using the PSCAD software to generate the different type of transient data caused by capacitor switching and lightning. Feature extraction is performed using wavelet technique. Next, the wavelet coefficients specifically the minimum and maximum values of the wavelet energy served as inputs for the SVM for classification purpose. Initial results showed that SVM is capable to classify the transient source with Radial Basis Function (RBF) as the kernel

Comparative analysis of input parameters using wavelet transform for voltage sag disturbance classification

Voltage sag is one of the major power quality disturbance today. Significant causes of voltage sag include power system fault, starting of induction motor and energizing of transformer. This paper focus on a comparative analysis carried out to determine the best input parameters to be employed in support vector machine (SVM) method for voltage sag cause classification. Wavelet transform is carried out to extract the feature of these voltage sags which are used as the input to the SVM. Two different types of inputs used, namely the energy level and the min-max values of the wavelet. Comparisons between these input parameters are made to determine the best method which give the best prediction values. Training and testing data based on the standard IEEE 30 bus distribution system are simulated using PSCAD software. The results show that the performance of the min-max wavelet values as the input to the SVM is superior than the energy level input in term of accuracy and computational time.

A New Method For Locating The Source Of Lightning Transients

Power system transients are power quality disturbances that can be harmful especially to electronic equipment. This paper presents a new method to locate the source of lightning transient relative to the monitoring point. The proposed method is based on complex wavelets energy which is calculated from the voltage and current data obtained at the monitoring point. The data is transformed to its complex wavelet counterpart. The transient power is first calculated and the complex wavelet energy is obtained by integrating the transient power. From the wavelet energy plot against time, a change in wavelet energy from an approximately zero to a negative value indicates that the transient source is from downstream or in front of the monitoring point. On the other hand, a change in wavelet energy from an approximately zero to a positive value indicates that the transient source is from upstream or behind the monitoring point. To verify the proposed complex wavelet method, simulations using the PSCAD/EMTDC software have been performed. Lightning strikes in a PSCAD library model have been adapted for the impulsive transient simulation. Simulation results prove that complex wavelet energy is capable of locating accurately the source of transients in a power distribution system.

Investigation on magnetic resonant coupling for low voltage application

This paper describes the characteristics of magnetic resonant coupling at 27.12 MHz for low voltage wireless power transfer system (WPT). Based on spiral magnetic inductance coupling, four single layer air coils are mutually coupled and parameterized to represent power coil, transmitter, receiver and load. The system was simulated using electromagnetic full wave simulator and analysis was performed to characterize the effect of coupling for maximum efficiency of power transfer.