Smart metering and energy access programs: an approach to energy poverty reduction in sub-Saharan Africa

  

Ethical clearance reference number: refer to the uploaded document Ethics Certificate.pdf.

General (0)

• 0 - Built diagrams and figures.pdf: diagrams and figures used for the thesis

Analysis of country data (1)

• 0 - Country selection.xlsx: In this analysis the sub-Saharan country (Niger) is selected based on the kWh per capita data obtained from sources such as the United Nations and the World Bank. Other data used from these sources includes household size and electricity access. Some household data was projected using linear regression. Sample sizes VS error margins were also analyzed for the selection of a smaller area within the country.

Smart metering experiment (2)

• The figures (PNG, JPG, PDF) include:

           - The experiment components and assembly

           - The use of device (meter and modem) softwar tools to program and analyse data

           - Phasor and meter detail

           - Extracted reports and graphs from the MDMS

• The datasets (CSV, XLSX) include:

           - Energy load profile and register data recorded by the smart meter and collected by both meter configuration and MDM applications.

           - Data collected also includes events, alarm and QoS data.

Data applicability to SEAP (3)

• 3 - Energy data and SEAP.pdf: as part of the Smart Metering VS SEAP framework analysis, a comparison between SEAP's data requirements, the applicable energy data to those requirements, the benefits, and the calculation of indicators where applicable.
• 3 - SEAP indicators.xlsx: as part of the Smart Metering VS SEAP framework analysis, the applicable calculation of indicators for SEAP's data requirements.

Load prediction by machine learning (4)

• The coding (IPYNB, PY, HTML, ZIP) shows the preparation and exploration of the energy data to train the machine learning model.
• The datasets (CSV, XLSX), sequentially named, are part of the process of extracting, transforming and loading the data into a machine learning algorithm, identifying the best regression model based on metrics, and predicting the data.

HRES analysis and optimization (5)

• The figures (PNG, JPG, PDF) include:

           - Household load, based on the energy data from the smart metering experiment and the machine learning exercise

           - Pre-defined/synthetic load, provided by the software when no external data (household load) is available, and

           - The HRES designed

           - Application-generated reports with the results of the analysis, for both best case HRES and fully renewable scenarios.

• The datasets (XLSX) include the 12-month input load for the simulation, and the input/output analysis and calculations.
• 5 - Gorou_Niger_20220529_v3.homer: software (Homer Pro) file with the simulated HRES

· Conferences (6)

• 6 – IEEE_MISTA_2022_paper_51.pdf: paper (research in progress) presented at the IEEE MISTA 2022 conference, occurred in March-2022, and published in the respective proceeding, 6 - IEEE_MISTA_2022_proceeding.pdf.
• 6 - ITAS_2023.pdf: paper (final research) recently presented at the ITAS 2023 conference in Doha, Qatar, in March-2023.
• 6 - Smart Energy Seminar 2023.pptx: PowerPoint slide version of the paper, recently presented at the Smart Energy Seminar held at CPUT, in March-2023.