Electrical & Electronics Engineering

DEVELOPMENT OF A DEEP CONVOLUTIONAL NEURAL NETWORKBASED MOBILE APPLICATION FOR DETECTING PNEUMONIA FROM CHESTX-RAY IMAGE

Mon, 01 Nov 2021 00:00:00 GMT

DEVELOPMENT OF A DEEP CONVOLUTIONAL NEURAL NETWORKBASED MOBILE APPLICATION FOR DETECTING PNEUMONIA FROM CHESTX-RAY IMAGE DADA, ADEFILA BOLUWAJI In rural communities, access to reliable healthcare is a major challenge. The rural people depend on traditional remedies to tackle different ailments. Hence, acute inflammation of the respiratory tract caused by viruses and bacteria could lead to a condition called pneumonia. This condition always proves difficult to manage with traditional medicine. Hence, there is a need to introduce a technology-driven solution for early detection of pneumonia. Thus, in this research, a Deep Convolutional Neural Network enabled mobile app for detecting pneumonia was developed and its performance was elaborately evaluated. This research was executed in five stages, namely Data Acquisition, Model Training, Development of Mobile Application, and Performance Evaluation. In the Data Acquisition stage, chest x-ray image dataset was acquired from Kaggle which was then preprocessed and stratified in the Data Preparation stage. The preprocessed and raw dataset were used to train a group of TensorFlow models in the Model Training stage. The models were then deployed to the mobile app during the Development of Mobile Application stage. In the final stage, which is the Performance Evaluation and Validation stage, the performances of the trained models and the mobile app was determined, and were subsequently validated by a trained radiologist. From the analysis of the results obtained, Strata 3 outperformed both Strata 1 and Strata 2, with accuracies of 95% and 97%, respectively, for the Normal and Pneumonia classes. The developed mobile app responded differently based on the data Strata used in training the TensorFlow model. The 2nd and 3rd Strata, respectively showed a response time of 93.3ms and 92.73ms. During the validation substage, an error rate of 13.3% was obtained for the 3rd Strata when validated by a trained radiologist while an error rate of 26.6% was obtained for both the 1st and 2nd Strata When validated by the mobile application.

AN INVESTIGATION INTO POWER QUALITY-BASED REINFORCEMENTS AND COSTS FOR SELECTED ELECTRICITY DISTRIBUTION NETWORKS

Sat, 01 Jan 2022 00:00:00 GMT

AN INVESTIGATION INTO POWER QUALITY-BASED REINFORCEMENTS AND COSTS FOR SELECTED ELECTRICITY DISTRIBUTION NETWORKS OLAKUNLE, OLUWADARE RAPHAEL The study aims to provide adequate power quality-based reinforcements for typical electricity distribution networks with the proliferation of non-linear loads and to determine its cost implication to the utility and the electricity users. In this study, a simulation approach was used in the analysis and reinforcements of IEEE 33-bus and Ajilosun 43-bus feeder networks using NEPLAN software. The load flow and harmonic flow module of NEPLAN software was used to investigate the voltage profile and harmonic distortion levels in the distribution networks. The results obtained showed that 81% and 86% of the 11 kV buses of the IEEE 33-Bus and Ajilosun 43-Bus distribution networks respectively violated the specified ± 5% voltage deviation limit while 53% and 79% of the 11 kV buses of the IEEE 33-Bus and Ajilosun 43-Bus distribution networks respectively violated the specified 5% harmonic distortion limits specified in the IEEE 519 standard. Hybrid optimization techniques that combine the Genetic Algorithm (GA) and Direct Search method were used in the sizing and location of capacitor bank and passive harmonic filters to achieve optimal values for the voltage profile and voltage Total Harmonic Distortion THDV level at the load buses. The obtained results showed that the voltage profile of the 11 kV load buses improved significantly to the range of 10.69 kV (97.19%) to 11.43 kV (103.88%) and 10.64 kV (96.73%) to 11.03 kV (100.27%) for the IEEE 33-bus 43-bus and Ajilosun 43-bus distribution network feeders respectively. The THDV propagation on the 11 kV network was reduced to the range of 2.13% to 3.18% for the IEEE 33-Bus network and 2.99% to 4.46% for Ajilosun 43-Bus feeder network. The investment cost analysis for the implementation of these reinforcements was estimated to determine the economic feasibility of the reinforcements’ investment. The estimated cost of reinforcements based on different reinforcement investment cost options was in the range of 35.19 million Naira to 53.81 million Naira and 48 million Naira to 138.81 million Naira for the IEEE 33-Bus and Ajilosun 43-Bus distribution networks respectively. A reinforcement billing charge of up to ₦20/kWh of energy consumption resulted in a payback period of within two years for the different reinforcement investment cost options for the IEEE 33-bus and Ajilosun 43-bus distribution networks. The study showed that the investment in power quality-based reinforcements of the distribution networks achieved good quality operating parameters and the costs recoverable within two years based on selected tariff on energy consumption.

MODELLING AND EVALUATION OF THE PROTECTIVE SYSTEM

Tue, 01 Jun 2021 00:00:00 GMT

MODELLING AND EVALUATION OF THE PROTECTIVE SYSTEM FALEYE, OLALEKAN PETER this research, modeling and evaluation of the protective system of Omu Aran/Otun 33 kV feeder network is considered and a search for the protective system improvement is obtained. System data of the feeder network were obtained to create the feeder network’s load flow analysis and short circuit analysis using Electrical Network Modeling and Simulation software tool called Electrical Transient Analyzer Program (ETAP). The software creates an electrical digital twin and analyzes electrical power system dynamics, transients and protection. In Nigeria, power supply availability, sufficiency and reliability are major operational challenges. In power system, sometimes faults occur and affect system reliability. Proper evaluation of the operations of the protective system is important to identification of design deficiency, component failure and incorrect settings which will improve the reliability of the feeder network. A short circuit analysis was performed in addition to the load flow analysis which tells the possible fault currents in the system. The three-phase, lineground (LG), line-line (LL) and line-line-ground (LLG) fault currents for a fault on each bus respectively. The load flow analysis was carried out to identify the conditions of the network while the short circuit analysis was carried out to determine the faults currents associated with the network. The load flow results: percentage voltage profile obtained from the load flow result shows that the voltage values at the buses fall within the permissible limit with the highest and lowest voltages of 100.06V and 98.07V respectively. Also, the percentage loading results as show that the Aaye Transformer is critical with percentage loading of 298.7%. The percentage voltage profile of the network under fault shows that the network cannot regain its stability under fault conditions. The network experience a worst and best voltage dip of 30.2% and 11.7% at Ilogbo and Imode Ekiti respectively.

DEVELOPMENT OF PATH LOSS MODEL FOR ORANGE FM RADIO STATION IN ONDO STATE, NIGERIA

Sun, 01 Aug 2021 00:00:00 GMT

DEVELOPMENT OF PATH LOSS MODEL FOR ORANGE FM RADIO STATION IN ONDO STATE, NIGERIA ALABI, KABIR The measurement of the electric field strength distribution of Orange FM radio station, Akure, Ondo State transmitting on 94.5 MHz was conducted round the stretch of the coverage areas. The latitude, longitude, elevation and line of sight distance to the transmitter base were measured at every access point at 4 km intervals from each other using Global Positioning System Receiver “Garmin Handheld Personal Navigator (GPS-72)”, while the field strengths were measured with CATV S110 digital signal level meter. Measurements were taken along four (4) major routes away from the transmitter base. The path loss values were calculated from the electric field strength measured in all the routes and the path losses were plotted against line of sight distances. For each route, least square regression analyses were carried out on each plot and a sub-model was developed therefrom. The four submodels were later averaged to form a single model for the considered routes. Comparative evaluation between the developed path loss model and two of the existing models (Egli and Free Space Path Loss (FSPL)) were carried out as well. Root Mean Square Error (RMSE) was calculated for each sub-model. From the RMSE results analyses, the Egli model generated lesser values (compared to the FSPL) which are within acceptable range (that is, RMSE < 6dB for all the routes). This showed that the developed model did not over-predict the path loss. The developed model will enable the broadcasting companies to determine the coverage of their transmitter and improve on the areas with poor signal reception. The information from this developed model prediction can be a useful tool for link budget design for Orange FM station in Akure, Ondo State. It will also provide a broad idea of path losses of FM signals to radio engineers operating in this environment.The model is therefore valid and can be used to estimate path loss of Orange FM 94.5 MHZ signals in Akure, Ondo State, Nigeria.