Electrical Computer Engineering

Permanent URI for this collectionhttps://etd.hu.edu.et/handle/123456789/74

Browse

Now showing 1 - 20 of 69

RELIABILITY ASSESSMENT AND ENHANCEMENT OF DISTRIBUTION SYSTEM BY INCORPORATING RENEWABLE DISTRIBUTED GENERATION
(Hawassa University, 2016-04-27) YISHAK KIFLE
Electricity is produced and delivered to customers through generation, transmission and distribution systems. A stable and reliable electric power supply system serve customer loads without interruptions. Distribution systems deliver power from bulk power systems to customers. Distribution reliability primarily relates to equipment outages and customer interruptions. This thesis work attempts to thoroughly identify causes for power interruptions and customer dissatisfaction and discusses the design, reliability and operation and maintenance of Hawassa feeder 10 distribution networks. With recent advances in technology, utilities expect to see increasing amounts of distributed generation (DG) on the distribution systems. Reliability worth is very important in power system planning and operation. Having a DG ensures reliability improvement and may be used to increase the reliability worth. This research presents the study of a radial distribution system and the impact of placing renewable energy DG (PV, Wind) and battery energy storage in order to increase the reliability worth. The reliability improvement is measured by different reliability indices that include SAIDI, CAIDI, EENS and ASAI. The analyzed and calculated distribution reliability indices values have been compared with standard benchmark values and comparison clearly indicates that Hawassa feeder 10 distribution systems is extremely unreliable. The values for SAIDI, SAIFI, CAIDI and ASAI are 240 minutes/year, 1.5 interruptions/customer, 123 minutes/year and 99.91% respectively in a USA standard, whereas the corresponding values for the existing distribution system of Hawassa feeder 10 are 5129.002 minutes/year, 47.5205 interruptions/customer, 107.94 minutes/year and 99.02% respectively. also unavailability of energy per year is 93.706mwh. The thesis also evaluates reliability of distribution networks, including islanded micro grid cases, is presented. The network includes two types of distributed energy resources solar photovoltaic (PV), wind turbine (WT) and energy storage as back up. These distributed generators contribute to supply part of the load during grid-connected mode, but supply the high densely loaded area Arbegona load during islanded micro grid operation. The studies performed are supported with the Power Management System Software ETAP and MATLAB CODED for power output of renewable energy source.
RESOURCE POTENTIAL ASSESSMENT AND OPTIMAL DESIGN OF PV/WIND/BATTERY/ DEISEL GENERATOR BACK-UP HYBRID ENERGY SYSTEM FOR REMOTE AREAS
(Hawassa University, 2016-10-22) MESFIN JARISO
Off grid hybrid systems have been attracting to supply electricity to rural areas in all aspects like, reliability, sustainability and environmental protections, especially for communities living far in areas where grid extension is not appropriate. Hybrid renewable set-up indicates that combinations based on the renewable sources could be applied simultaneously to supply energy in the form employed in an off-gird supporting with battery storage and diesel generator as backup systems. In this thesis photovoltaic-wind turbine-battery bank and diesel generator have been optimally sized, simulated and optimized for the rural community of Addis Boder village in the southern regional State, Ethiopia. Primary load demand of 142KWh/day, peak load of 26kW, deferrable energy is about 27kWh/day, and deferrable peak load of 3.6kW was involved during optimization of the power system. Well known HOMER modeling tool have been used to design the off-grid system. Solar and Wind energy are considered as primary sources to supply electricity directly to the load and to charge battery bank when excess generation is happened however in peak demand times diesel generator could also be engaged. Regarding solar energy potential there is no accurately recorded solar radiation database in the country, instead only sunshine hour data was available. Empirical formulas are used to estimate the solar radiation from available sunshine duration data. This result is compared with the data collected from NASA and it is found to be nearly the same. Consequently, for this study the calculated solar radiation is used for modeling the hybrid system. During the design of this power system set-up, the simulation and optimization was done based on the electricity load, climatic data sources, economics of the power components and other parameters in which the NPC has to be minimized to select an economic feasible power system. The results obtained from the software give numerous alternatives feasible hybrid systems with different levels of renewable resources penetration which their choice is restricted by changing the net present cost of each set up Power schemes with less NPC, less COE, higher renewable fraction, less capacity shortage, smaller excess electricity and minimum fuel consumption would be suggested as optimum system. It is concluded that besides electricity provisions, the role of a standalone hybrid system in protecting the environment from degradation, the improvement of life of people living in rural area, development of clean energy, and the future situation regarding fossil fuel sources should be taken in to account. Taking these issues into account the free solar and wind energy of the country should be utilized to improve the quality of life of the communities living in rural areas
Reactive Power Compensation and Harmonic Mitigation of Power Distribution Network Using Shunt Active Power Filter Based P-Q theory (Case study: Hawassa substation-2)
(Hawassa University, 2017-04-09) Tahir Kedir
Traditionally, the harmonic distortion and reactive power has been reduced using passive LC filters. However, the application of passive LC filters for reactive power compensation and harmonic mitigation results in parallel resonances with the network impedance. It is poor flexible for dynamic compensation of different frequency harmonic components.This work presented a scheme for reactive power compensation and harmonic mitigation of Hawassa substation-2 using shunt active power filter(SAPF), which is based on instantaneous p-q theory. The filter is identified, it automatically adapts to changes in the network and load fluctuations to mitigate the harmonics and reactive power issues. It is able to produce almost unity power factor and mitigate the total harmonics distortion (THD) to meet specified power quality standards. The performance of the SAPF depends mainly on the p-q theory to compute reference current and hysteresis band current controller to generate pulses for switching pattern of the inverter. The advantage of instantaneous p-q theory shows that it is instantaneous and works in time domain. MATLAB/Simulink computer simulation (version R2016a) is used as a simulation tool for this study. The simulation results before and after the compensation is compared to specified IEEE 519-1992 power quality standards.In three phase, three wire configurations, the result shows that THD in the load current is reduced from 37.46% to 3.56% and THD in the source voltage is reduced from 26.91% to 2.74% and power factor is improved from 0.869 to 0.949 under non ideal voltage source conditions while the reactive power is reduced from 1,178.04KVAr to 746.05KVAr
OPTIMAL PLACEMENT OF PHASOR MEASURMENT UNIT FOR STATE ESTIMATION IMPROVEMENT
(Hawassa University, 2017-10-15) KINFE NEGASH
Installation of modern sensing devices is an integral requirement in the transition of a conventional power grid to smart grid. Phasor measurements synchronized through GPS presents one such alternative to enhance the monitoring and control feature in the power grid. Due to the high financial implications, these units should be deployed optimally in the grid. The methods reported in the literature for optimal placement usually provide multiple solutions. This thesis presents criteria of optimal solution for state estimation enhancement. The criterion is based on sensitivity analysis and analyzing way of increasing the rank of the gain matrix to install the phasor measurement unit on improving the state estimation results. A reduced form integer linear Programing algorithm is to be formulated according the criterion given for optimal placement problem for improved state estimation. The reason behind this is that all the downstream applications of contingency analysis; steady-state, voltage, and dynamic as well as the optimal power flow are dependent on the accuracy of the state estimation results. Because of its key role in energy management system, state estimation is set to be a vital function in the development of smart grid. For the Process Power education tool (PET) and Matlab are used and validated on testing on IEEE 14-bus system as a base case and a southern region 21-bus system as a study case
Performance Analysis of Modulation Techniques for 5G Networks
(Hawassa University, 2018-03-17) GETACHEW HAGOS
The new 5G cellular network systems will feature several innovative strategies as compared to existing LTE systems, such as extensive adoption of small cells, use of mm-wave communications for short-range links, large scale antenna arrays installed on macro base stations, cloud-based radio access network, cognitive radio etc. All these strategies will be impacted by the modulation format used at the physical layer. At the same time, 5G cellular networks will have more stringent requirements than LTE in terms of latency, energy efficiency and data rates, which again are impacted by the adopted modulation scheme. Hence, proper selection and utilizing efficient modulation scheme is critical. The current modulation technique such as orthogonal frequency division multiplexing (OFDM) suffers from high peak to average power ratio which results in low efficiency of power amplifier and increases the battery consumption. Moreover, the OFDM spectrum has high out of band side lobes or side lobe leakage causing problem of low spectral efficiency. Therefore, OFDM can fulfill the requirements of 5G wireless networks in a limited way. In order to overcome some of these drawbacks of OFDM, new modulation techniques for 5G communication systems are considered. Among them, Filter bank multicarrier (FBMC), and Universal filtered multicarrier (UFMC) are expected be in race to be considered for the selection. In this paper, we perform the performance analysis of OFDM, UFMC, and FBMC in terms of its spectral efficiency (SE), power spectral density (PSD), peak to average power ratio (PAPR), and bit error rate (BER) using the MATLAB software tools. We analyze the results by varying the parameters of the modulation techniques which affects the performance of OFDM, UFMC, and FBMC. The plot shows that, FBMC has better SE for large burst durations but for small burst duration the SE of UFMC is better. In terms of PSD and BER FBMC is better. In terms of PAPR UFMC is better. The result of this research will play a significant role in selecting efficient modulation scheme for the upcoming 5G cellular networks
IMPROVING VOLTAGE SAG AND VOLTAGE SWELL OF DISTRIBUTION SYSTEM WITH INTEGRATION OF ULTRA CAPACITOR AND DYNAMIC VOLTAGE RESTORER
(Hawassa University, 2018-08-26) MEKONNEN SOLOMON
Power Quality (PQ) problem is an occurrence manifested as a nonstandard voltage, current or frequency that results in a failure of end use equipments. The supply status of electrical services required for case study of Agro Processing Industry illustrates this concept. The voltage sag and swell are the most frequent PQ problems that mainly occur in the distribution systems because of it causes circuit breaker tripping, failure of drive systems, shutdown for domestic and industrial equipment. The Dynamic Voltage Restorer (DVR) connected in series has magnificent dynamic capabilities and is a flexible solution for PQ problems. Ultra-Capacitors (UCAP) has ideal characteristics such as high power and low energy density essential for the mitigation of voltage sag and swell. In this proposed research voltage sag and voltage swell problem are improved by using the method of integration of Ultra Capacitor and Dynamic Voltage Restorer device. In this study UCAP was used as energy storage as it provides excessive power in a short time interval of time. Integrated DVR into Ultra Capacitor via bidirectional DC-DC converter which supports in presenting a rigid dc-link voltage and helps in compensating temporary voltage sag and voltage swell. The integrated UCAP-DVR is implemented at low voltage side of distribution transformer. PI Controller was used in DVR for power quality enhancement. The simulation results carried out by Matlab/Simulink verify the performance of the proposed method. In the proposed system the voltage sag is mitigated from 0.304 p.u to 1p.u and the voltage swell is compensated from 1.125p.u to 1p.u
Peak Load Forecasting on a Case Study of Ethiopian Electricity Demand
(Hawassa University, 2018-10-27) Abebaw Gashaw
Load forecasting is a way of estimating what future electric load will be for a given forecast horizon based on the available information. Peak load forecasting is the procedure to predict future peak loading conditions. This thesis work is a study about Ethiopian peak load forecasting for the next 10 years. In this thesis Hybrid method used for the peak load forecasting, the method is the combination of two models: (i) econometric model, which uses the multiple regression techniques and (ii) end use model which use the data directly from the user demand. The econometric approach combines economic theory and statistical techniques for forecasting. The approach estimates the relationship between peak load (dependent variables) and load influencing factor. In this thesis the economic and demographic factors are considered. In other word Gross Domestic Product (GDP) of Ethiopia and population of Ethiopia are the selected factors. The econometric approach combines economic theory and statistical techniques for forecasting electricity demand. In the econometric model multiple regression technique is used for mathematical analysis. This analysis is further compared with ARMAX mathematical model. A raw data regarding load demand covering previous period of five years data is used for this study. In this work peak load is used. The GDP and population of previous data with load data are used to get coefficients .The Mean Forecasting Error (MFE) also calculated to reduce the error. Using this coefficients and error in the mathematical equations, the next ten years forecast result is found. The study also utilized c-sharp software and forecasted the Ethiopian ten years peak load demand and compared with the econometric method load forecasted data. End-use model focuses on the various uses of electricity in the users. These models are based on the principle that electricity demand is derived from customer demand. Thus end-use model explains load demand as a function of the number of users. I In this thesis the newly commenced load consumed sector is independently analyzed and data collected. The end-use modeling, econometric modeling, and their combinations are the most often used methods for long-term load forecasting. Long-term forecasts include the forecasts on the population changes, economic development, industrial construction, and technology development. Here finally this thesis combines the two models and the 10 years forecasted result found by adding the models result
PERFORMANCE EVALUATION OF INTERFERENCE MANAGEMENT STRATEGIES IN 5G HETEROGENEOUS NETWORKS
(Hawassa University, 2019-04-13) ELIAS MAMUSHET MENGESHA
In the rapid growth of the internet, connected mobile devices need ultra-high-speed and high definition multimedia services. In order to meet the demands of the future networks, International Mobile Telecommunication system for 2020 (IMT-2020) indicates that the wireless spectrum resource requirements for 5G wireless networks are primarily a result of the growing demand for system capacity. Due to the increase of mobile data traffic, heterogeneous networks (HetNets) are the promising approach to handle the demand of the recent mobile traffic load. The development of HetNets facilitates subscribers Quality of Experience (QoE) guarantee and higher data rate. However, the interference management is one of the challenges in HetNets due to the coexistence of multiple small cells. In this thesis focus on the capacity and reliability improvement investigation of HetNets deployment using stochastic geometry base station deployment. The simulation result shows that the smaller user densities have higher per-user capacity due to the sufficient available channel resources relative to the user density and smaller interference is resulting from the smaller quantity that a given sub-channel is occupied. As the number of subscribers increases the per-user capacity decreases because the available spectrum resource is allocated to each user. The increment in Pico Base Station (PBS) density causes more interference between picocells and when PBS density is high enough, the receivers in picocells severely suffer interference from neighboring receivers. The area spectral efficiency of picocells is lower while the MU power is high. The outage probability of both uplink and downlink transmissions are increasing as PBS density increases. This is due to the fact that high PBS density causes more interference to the overall network and with higher MU density, the outage probabilities for picocells is high. Finally, when the number of cooperating BSs and transmit power of BSs increases the energy efficiency of the HetNets is decreased while the number of cooperating BSs and transmit power of BSs increases the spectral efficiency of the HetNets is increased
OPTIMAL EXPANSION PLANNING OF DISTRIBUTION NETWORK WITH DISTRIBUTED GENERATION BY UTILIZING GRID-BASED MULTI-OBJECTIVE HARMONY SEARCH ALGORITHM (CASE STUDY: DEBREMARKOS DISTRIBUTION NETWORK)
(Hawassa University, 2019-04-25) TAKELE FEREDE AGAJIE
Electrical energy plays a vital role in the socio-economic development. To combat for the power system profile problems, distribution substation needs to be established considering future expansion due to urbanization. Debre Markos (D/M) distribution network needs expansion planning to meet the growing load demand. To evaluate the capability of the existing distribution network and to supply reliable power for future expansion, demand forecast for the years 2017/18-2022/23-2027/28 has been performed by using trend forcasting technigue withl east sequre approximation and evaluating the load flow by using backward-forward sweep load flow. According to the results, the existingnetwork cannot meet the existing load demand and it has major problems of increased voltage deviation and power loss. In this thesis, D/M distribution network expansion planning considering future demand growth and distributed generation placement and sizing is carried out using Grid based Multi-Objective Harmony Search Algorithm (GrMHSA). The total real power loss (Pl), total reactive power loss (Ql) and total voltage deviation (VD) at the target year for the base case by taking the existing line and the projected bus data are 7434.9kw, 7391.8kvar and 58.6952p.u for D/M Feeder 3 and 470.7058kw, 404.5524kvar and 6.4412p.u for D/M Feeder 4 respectively. After applying GrMHSA optimization technique for DG sizing and placement, the total Pl, total Ql and VD at the target year are 95.398kw,124.979kvar and 0.479p.u for D/M Feeder 3 (F3A and F3B) and 30.811kw, 37.727kvar and 0.533p.u for D/M Feeder 4 respectively
ENHANCING LOW VOLTAGE RIDE THROUGH CAPABILITY OF DFIG BASED WIND TURBINE: A CASE STUDY OF ADAMA-II WIND TURBINE
(Hawassa University, 2019-08-18) FSAHA MEBRAHTU
Doubly fed induction generators (DFIG) based wind turbines (WT) are major sources of energy generation around the world. One of such energy source is wind generation, which is very much popular now days. It utilizes doubly fed induction generator (DFIG) to generate electricity in generation mode. But sometimes wind turbines are disconnected from the power system in the event of grid faults. To avoid grid instability as well as to ensure the secure and reliable operation of the power systems, all wind turbines must stay connected with the grid during faults and capable to ensure the operation of the utility grid side during voltage sags in the event of a fault that is performed by analyzing the low voltage ride through (LVRT) capability of DFIG based wind turbines.. DFIG is very sensitive to any grid disturbances, if a severe voltage dip occurred due to grid fault, high currents will pass through stator and rotor windings and also a very high DC voltage would be induced in converter circuit, which may lead to damage the converter circuit and the DFIG windings. A crowbar protections system is essential to avoid the disconnection of the doubly fed induction wind generators from the network during faults and to protect the DFIG from the dangerous effects of the electrical fault. Accordingly, Ethiopia have been started to expand its energy generation by installing renewable energy source like wind farm. Presently, Ethiopia has installed wind farms like Ashegoda, Adama-I and Adama-II. The Adama-II wind farm project has 153 MW with 102 units, each of with generation capacity of 1.5 MW. This work focused on the improvement of low voltage ride through capability (LVRT) of Adama-II wind turbines. In this thesis a crowbar protection system is used to improve the power system fault ride through capability of DFIG of Adama-II. During the symmetrical fault using crowbar protection the value of crowbar current at time three 3 seconds is overshoot to 3000 A and it will reduce to the original condition after 100 ms. Also the value of stator voltage, stator flux during faulty and symmetrical conditions before the fault and after are the same, but during the voltage dip the stator flux is decreased and the stator voltage is also reduced. The dynamic behavior of DFIG under normal and various grid fault conditions are simulated and analyzed to provide the recommendations. MATLAB/Simulink software is utilized for the development of proposed system
PERFORMANCE COMPARISION OF PILOT-BASED CHANNEL ESTIMATION METHOD IN OFDM SYSTEM OVER FADING CHANNEL
(Hawassa University, 2019-08-20) HABTAMU ABEBE
Wireless channel is unknown and inclined to time dispersion, which causes inter symbol interference (ISI) and fading. OFDM is one of the most capable core technologies in the fourth generation of wireless communication systems. OFDM has the advantage over the conventional single-carrier modulation schemes when the channel is frequency-selective fading and widely used in current wireless networks. In this thesis work the performance of two pilot-based channel estimation algorithms; namely, least square (LS) and Minimum Mean Square Error (MMSE) with corresponding channel interpolation technique is investigated for digital communication systems. Channel estimation is a technique that can be used to design a channel in a particular surrounding based on known train symbols which is inserted in the input of transmitter end and tested at the receiver end. This thesis performs , analysis of channel estimation algorithms based on bit error rate (BER) and signal to noise ratio (SNR) minimizing the error distance between the transmitted and received data. To estimate the channel coefficients; The MATLAB plot can perform for QAM, 16QAM, and 64QAM constellations. For different international telecommunication union (ITU) channel models Such as, the portable outdoor (PO), portable indoor (PI), Rural Area six path (RA6), Typical Urban six path (TU6), and Rayleigh channel models. Typical urban channel model has better at higher order modulation. From the MATLAB results, LS and MMSE algorithms are compared based on their pilot ratio and channel length. LS estimation has better at high SNR value compared to MMSE estimation
PERFORMANCE COMPARISON OF CHANNEL MODELS FOR UFMC SYSTEM
(Hawassa University, 2019-08-29) TILAHUN BELAYNEH MAMO
This paper presents the bit error rate (BER) vs. signal-to-noise (SNR) ratio performance of channel models under UFMC system with 32- FSK, PAM, DPSK, 4-QAM, 16-QAM, 32-QAM, 64-QAM and 128-QAM modulation. In future it is believed to be more challenging and complicated with the evolution of different types of fading model. These types of fading models determined in urban and rural areas. Rayleigh fading model is considered to be most common fading model, found in urban environment for non-line of sight. Rician fading consists of line of sight communication and found to be more applicable for satellite communication. Nakagami fading model is mostly suited for urban multipath propagation and it is sought to be most practical model, specially used in mobile communication. In order to achieve error free signal at the receiver the choice of modulation scheme should be done wisely. The combination of best efficient modulation scheme along with the block coding helps in getting the signal error free at the receiver. The condition of signal is decided by SNR v/s BER (bit error rate) simulation. Our main objective is to get error free result at the receiver, so for this different data transmission techniques are used to transmit the data from transmitter to receiver in various fading channels under different modulation schemes. Effect of shape factor on Nakagami fading is also covered in this thesis to get efficient results. The BER performance was to compared with Additive White Gaussian Noise (AWGN), Rayleigh-faded, Nakagami-m faded and Rician-faded channels. It has been further concluded that the BER vs. SNR performance graph in AWGN channel environment is better than that in Rayleigh, Nakagami-m and Rician-faded channel. It is also observed that by varying the standard-deviation of the channel, the BER vs. SNR performance graph is not going to be affected if the channel is considered Rayleigh-faded. In this paper we address a new aspect of UFMC system performance by investigating its capability to overcome the channel fading channels’ effects and to provide good and reliable performance levels overcome these channel effects.
PERFORMANCE COMPARISION OF PILOT-BASED CHANNEL ESTIMATION METHOD IN OFDM SYSTEM OVER FADING CHANNEL
(Hawassa University, 2019-10-11) HABTAMU ABEBE
Wireless channel is unknown and inclined to time dispersion, which causes inter symbol interference (ISI) and fading. OFDM is one of the most capable core technologies in the fourth generation of wireless communication systems. OFDM has the advantage over the conventional single-carrier modulation schemes when the channel is frequency-selective fading and widely used in current wireless networks. In this thesis work the performance of two pilot-based channel estimation algorithms; namely, least square (LS) and Minimum Mean Square Error (MMSE) with corresponding channel interpolation technique is investigated for digital communication systems. Channel estimation is a technique that can be used to design a channel in a particular surrounding based on known train symbols which is inserted in the input of transmitter end and tested at the receiver end. This thesis performs , analysis of channel estimation algorithms based on bit error rate (BER) and signal to noise ratio (SNR) minimizing the error distance between the transmitted and received data. To estimate the channel coefficients; The MATLAB plot can perform for QAM, 16QAM, and 64QAM constellations. For different international telecommunication union (ITU) channel models Such as, the portable outdoor (PO), portable indoor (PI), Rural Area six path (RA6), Typical Urban six path (TU6), and Rayleigh channel models. Typical urban channel model has better at higher order modulation. From the MATLAB results, LS and MMSE algorithms are compared based on their pilot ratio and channel length. LS estimation has better at high SNR value compared to MMSE estimation.
ASSESSMENT OF RELIABILITY ASPECTS IN POWER DISTRIBUTION SYSTEM AND DEVELOPMENT OF REINFORCEMENT MECHANISMS (CASE STUDY: SHASHEMENE CITY DISTRIBUTION SYSTEM
(Hawassa University, 2019-10-17) KABTAMU KETSELA
A stable and reliable electric power supply system is an inevitable pre-requisite for the technological and economic growth of any nation. Due to this, Ethiopian electric utility (EEU) must strive and ensure that the customers’ reliability requirements, which are met at the lowest possible cost. EEU should have their own reliability improvement strategy depending upon their needs and requirements of their customers. Although there are notcomplete methods in managing all reliability issues, utilities can maximize network performance and better serve customers by carefully addressing trouble prone areas. In order to achieve this objective, reliability factors based reliability analysis is performed under Electrical Power System Analysis (ETAP) software environmentIn this thesis, it is observed that different reliability reinforcement mechanismshas been undertaken to evaluate predictive reliability analysis for the future system. Using line reconductoring for wendogent feeder the SAIFI, SAIDI and EIC reduced by 12%, 7.5% and 32% respectively. For shashemene city feeder using Autoreclosure SAIFI, SAIDI and EIC reduced by 45%, 35% and 35 % respectively. For Arsinegele feeder using separate CB the SAIFI, SAIDI and EIC reduced by 9.7%,6.4% and 8.8% respectively and water pump feeder using covered conductor SAIFI, SAIDI and EIC reduced by 14.9%,9% and 9.8% respectively. Reliability indices of shashemene substation show the substation is unreliable as compared to standard practices and Ethiopia’s reliability requirement
PSO Based Reactive Power Optimization of Utility Grid With Wind Generation
(Hawassa University, 2019-10-22) HAILU TIBEBE MENGISTU
Reactive power is critical to the operation of the power systems on both safety and economic aspects. Irrational distribution of the reactive power would severely disturb the power quality of the power systems and increases the transmission loss. Reactive Power Optimization (RPO) is defined as the minimization of active power loss by controlling a number of variables. Reactive power optimization problem is a nonlinear optimization problem and has both equality and inequality constraints. In this thesis, the Particle Swarm Optimization (PSO) algorithm and Mat-power toolbox are used to solve the reactive power optimization problem for generation/distribution system with wind generation (WG) plant. The proposed method minimizes the active power loss in a practical power system as well as determines the optimal placement of a new installed WG (Wind Generation). The practical 21-bus, power distribution network of Southern Region is used to evaluate the performance.
OPTIMAL SIZING AND PLACEMENT OF CAPACITOR AND DISTRIBUTED GENERATION FOR LOSS MINIMIZATION IN UNBALANCED DISTRIBUTION NETWORK
(Hawassa University, 2019-10-22) Bawoke Simachew
Power loss reduction is an important problem that needs to be addressed as generating electrical power. It is important to reduce power loss using locally generated power sources and/or compensations. This thesis presents a method of maximizing energy utilization, feeder loss reduction and voltage profile improvement for radial distribution network using the active and reactive power sources. Distributed Generation (DG) (wind and solar with backup by biomass generation) and shunt capacitor (QG) for reactive power demand are used. Integrating DG and QG at each bus might reduce the loss but it is economically unaffordable for developing countries like Ethiopia. Therefore, this work utilized an optimization method namely Whale Optimization Algorithm (WOA) for finding an optimal size and location at feeder for placing QG and DG, so that the feeder loss is minimized. WOA is a meta-heuristic optimization derived from natural food hunting behavior of biggest mammal fish called whale. To see the performance WOA, it is compared with particle swarm optimization (PSO). In the process of optimization, the feeder carrying capacity is considered as the primary constraint to get the best minimum loss while voltage, position limits and the sum of equality constraints are kept considered. The performance of the applied method is performed on 35 and 40 bus feeders of Bahir Dar distribution network. From the results, the power loss had reduced from 339.5703 kW to 22 kW in Ghion feeder and from 126.2149 kW to 22.4 kW in Bata feeder using WOA. It also had reduced from 339.5703 kW to 27 kW in Ghion and from 126.2149 kW to 51.3 kW in Bata feeder using PSO. Hence, it can be noted that that WOA is superior to PSO in terms of loss minimization
Integrated energy optimization of smart home appliances with cost-effective energy management system (Case Study: Mid-Size Residential Home at Hawassa City
(Hawassa University, 2019-10-27) Tesfahun Molla
With the development of smart grid technology, residents have the opportunity to schedule their power consumption pattern in their home by themselves for the purpose of minimizing electricity expense, reducing the peak-to-average ratio (PAR) and peak load demand. The two way flow of information between electric utilities and consumers in smart grid opened new areas of applications. In this thesis, a general architecture of home energy management system (HEMS) would be introduced in a home area network (HAN) based on the smart grid scenario and then an efficient scheduling method for home power usage would be proposed. The energy management controller (EMC) receives the demand response (DR) information indicating the Time-of use electricity price (TOUP) through the home gateway (HG). With the DR signal, the EMC achieves an optimal power scheduling scheme that can be delivered to each electric appliance by the HG. The rooftop photovoltaic system used as local generation micro grid in the home and can be integrated to the national grid. These results in reduction of electricity bill and when solar energy production is more than the demand of the home, the excess energy could be exported to the national grid. Accordingly, all appliances in the home operate automatically in the most cost-effective way. Detail explanation and discussion of four scheduling techniques would be designed under TOUP. To calculate the fitness function; two powerful swarm intelligence based Meta-heuristic algorithms i.e. Grey wolf optimizer (GWO) and Particle swarm optimization (PSO) algorithms had been used. Finally, detail comparison and analysis between proposed methods and algorithms could be carried out. To display the result, MAT LAB 16(a) software environment would be used
EVALUATING CRYPTOGRAPHIC ALGORITHMS OF AUDIO AND VIDEO DATA FOR SYSTEM RESOURCE CONSTRAINED OF IoT DEVICES
(Hawassa University, 2019-10-27) WAGARI BERHANU
We are living in an internet period in which digital data exchange in electronic way has progressed tremendously and securing information has become a challenge. Unless a creative means of securing information has been developed, the problem will be vast and limitless. In a bid to keep the pace of the era, cryptography plays an important role in information security systems. Cryptography is a process of creating information indecipherable to an unauthorized person. It is quite a well-known fact that IoT (Internet of Things) is a system of interrelated computing devices or digital machines is the ability to transfer data over a network. the devices that used in IoT is low power devices for this reason, the system resource is one sever issues. To do so various cryptographic algorithms used in IoT devices. Meanwhile, these algorithms consume a significant amount of computing resources such as CPU time, memory and computation time. This thesis provides evaluating AES, Blowfish, RSA, Trivium and Elliptic Curve Cryptosystems based on system resource usage to securing IoT device communications on the basis of encryption time, decryption time, throughput, CPU time, power efficiency and memory usage in audio and video data using Java as the programming language to develop algorithms. The result obtained shows that AES has better performance with encryption time, decryption time, throughput rate, CPU time, power efficiency and memory usage with audio and video files in various format
A MICROGRID ENERGY MANAGEMENT SYSTEM IMPLEMENTSTION USING MIXED-INTEGER LINEAR PROGRAMMING
(Hawassa University, 2020-04-27) SINTAYEHU TERBUSH
The ever continued growth and development of distributed generation (DG) in the electrical grid system led to the increasing expansion of microgrids across the world. Microgrid is distributed power generation units, energy storage devices, and controllable loads with the capability to operate in both grid-connected and island modes. Microgrid's economic operation is achieved through an energy management system that optimally schedules distributed generations and storage devices and continuously balances supply and demand. In this paper, a formulation of optimal unit commitment and dispatch scheduling of DGs in a grid-connected microgrid is presented. Mixed-integer linear programming (MILP) is used to carry out the optimal resource scheduling model. The goal is to reduce the overall operating cost of the system by optimally utilizing an energy storage device and a diesel generator unit by using load and renewable energy generation prediction. Operational constraints such as generation limits of DGs, battery charging/discharging limits, and state of charge (SOC) limits to be satisfied during all intervals of operation. Simulation results indicate that the operational cost of the system is effectively reduced through optimal scheduling of an energy storage system and a diesel Genset unit using the proposed strategy
SENSOR-LESS FIELD ORIENTED SPEED CONTROL OF INTERIOR PERMANENT MAGNET SYNCHRONOUS MOTOR DRIVES USING MRAS OBSERVER AND NEURAL ADAPTIVE CONTROLLER
(Hawassa University, 2020-08-14) GECHANU ABEBE DURBE
In this thesis, develops a Model Reference Adaptive System (MRAS) is used to estimate speed for interior permanent magnet synchronous motor (PMSM) using Neural Network (ANN) .The algorithm is based on field oriented control theory. The proposed control scheme was a model reference adaptive system (MRAS) speed observer based on stator currents with space vector pulse width modulation technique. The objective of the proposed sensor-less control is to improve the speed control performance and robustness of PMSM drive under load variations. For speed estimation performed at low speed, pi-radial basis function neural network observer is proposed instead of the traditionally used pi controller in adaptation mechanism of the MRAS.In this work, the traditionally used pi-controller in adaptation mechanism of the MRAS observer is replaced by pi-radial basis function neural network observer. The improvement in Speed response during load variations is shown by comparative study between conventional PI and neural PI. Simulation result showed the performance of speed estimator under various operating conditions using MATLAB/SIMULINK environment