Faculty of Engineering and Built Environment
Permanent URI for this communityhttp://ir-dev.dut.ac.za/handle/10321/9
Browse
Search Results
Item Application of AI for frequency normalization of solar PV-thermal electrical power system(IEEE, 2020-08) Estrice, Milton; Sharma, Gulshan; Akindeji, Kayode Timothy; Davidson, Innocent EwaenGrid-connected solar-PV schemes have become a significant part of the energy balance in the power system to satisfy the growing request for clean, affordable energy. This study attempts to link solar-PV generation with conventional thermal power plants and to integrate the control zone resulting in a hybrid solar PV-thermal electric power system using an AC tie line. An analysis of the frequency dynamics for varying load conditions of the interconnected system is studied. Diverse approaches of proportional, integral, and proportional-integral fuzzy logic built controllers are design and tested in order to match the electric power with variable loads of the system and hence to normalize the frequency ofthe system in shortest possible time. A comparative analysis of the design topologies is conducted out for the PV-Thermal scheme. Results obtain from the implementation are shown to justify the performance of proposed control efforts, using MATLAB software toolItem Curbing electricity theft using wireless technique with communication constraints(IEEE, 2020-08) Tshikomba, Salome C.; Estrice, Milton; Ojo, Evans E.; Davidson, Innocent E.Utility services are experiencing a common problem of power losses, which impose a significant impact on their annual budget. Practically, power losses consist of technical losses and non-technical losses. Technical losses are due to operations and aging of infrastructure, while nontechnical losses (NTL) are due to non-metered energy. The focus is on managing non-technical losses using an automation wireless method. The wireless ZigBee technique is proposed and further investigated for communication failure over long distances while solving the problem of stealing electricity. Advance-metering infrastructure (AMI) technique and smart meters are feasible for system integration; that is why they are chosen to be part of this study. The success of the study depends on quality data of the Utility, meaning the more accurate the data, the easier the analysis of outliers. The operation and planning of revenue protection contain a large amount of data that needs to be worked on, so data mining assists in that regard. Then the load profiling method assists in illustrating the variation in demandJelectricalload over a specific time. This is a preliminary investigation using a wireless communication technique as a viable solution in curbing electricity theft. The uniqueness of the proposed ZigBee system is that it recognizes the everyday act of stealing electricity through tempering with the meter box and tapping of the supply.