174 lines
16 KiB
TeX
174 lines
16 KiB
TeX
|
%%
|
|||
|
|
%% Created in 2018 by Martin Slapak
|
|||
|
|
%% last update: 2021-09-21
|
|||
|
|
%%
|
|||
|
|
%% Based on file for NRP report LaTeX class by Vit Zyka (2008)
|
|||
|
|
%% enhanced for MI-MVI (2018) and tuned for BI-PYT (2020, 2021)
|
|||
|
|
%%
|
|||
|
|
|
|||
|
|
\documentclass[english]{article}
|
|||
|
|
|
|||
|
|
\usepackage[utf8]{inputenc}
|
|||
|
|
\usepackage{array}
|
|||
|
|
\usepackage{float}
|
|||
|
|
\usepackage{hyperref}
|
|||
|
|
\usepackage{graphicx}
|
|||
|
|
\usepackage{booktabs}
|
|||
|
|
\usepackage{tabularx}
|
|||
|
|
\graphicspath{ {fig/} }
|
|||
|
|
\usepackage{color}
|
|||
|
|
\usepackage{lmodern}
|
|||
|
|
\usepackage[style=iso-numeric]{biblatex}
|
|||
|
|
\addbibresource{reference.bib}
|
|||
|
|
\title{Renewable energy as a driving factor in access to electricity in Kenya}
|
|||
|
|
|
|||
|
|
\author{Štěpán Beran \\ FIT ČVUT \\ beranst6@fit.cvut.cz}
|
|||
|
|
|
|||
|
|
\def\file#1{{\tt#1}}
|
|||
|
|
|
|||
|
|
\begin{document}
|
|||
|
|
\maketitle
|
|||
|
|
|
|||
|
|
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
|||
|
|
\section{Abstract}
|
|||
|
|
|
|||
|
|
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
|||
|
|
\section{Introduction}
|
|||
|
|
The tropical rain belt, located around the equator, boasts the greatest solar potential on Earth. This climate zone is home to the Amazonian rainforest and provides a fertile ground for permaculture farms. These farms are able to grow food year-round at competitive prices, without the need for costly heating systems, unlike farms in temperate regions like Spain and the Netherlands, which supply much of Europe's food. However, many countries in this tropical zone, including Kenya, largely rely on fossil fuels (especially diesel generators) as their main source of electricity in rural areas. These rural areas are often the most suited for permaculture, as opposed to urban areas, which tend to receive more energy infrastructure investment.
|
|||
|
|
|
|||
|
|
Kenya’s electricity mix has undergone significant transformation over the past two decades, shifting from a fossil-fuel-based system to one that is nearly entirely renewable. This change has been largely driven by public and private investments in geothermal and hydroelectric power plants. The costs of solar energy have also dropped drastically, with solar panels now costing about 95~\% less than they did two decades ago (calculated price per watt in 2000 compared to 2024). While the initial capital cost of solar energy systems may still be a barrier, their operational costs are far lower than that of diesel generators, which require continuous and costly fuel supplies. As described later in the theoretical part, the cost of installation of a stand-alone solar system is cheaper than connecting to the grid \cite{moner-girona-electrification}.
|
|||
|
|
|
|||
|
|
The high upfront costs of solar technology are increasingly supported by governmental initiatives such as the Kenya Off-Grid Solar Access Project (KOSAP), which has been funded by a World Bank loan of \$150 million USD \cite{kosap-wb}. This initiative aims to provide electricity to millions of Kenyans who currently lack access. The KOSAP program focuses not only on solar home systems and clean cooking solutions but also on the creation of "mini-grids" for community facilities, enterprises, and solar-powered water pumps\cite{kosap}.
|
|||
|
|
|
|||
|
|
This paper analyzes the role of renewable energy—such as geothermal, wind, and solar energy—as a key driver in increasing electricity access in Kenya. Traditionally, the push for renewable energy has been motivated by the climate conservation goals of developed nations, which have provided the resources and funding needed to stimulate demand and drive innovation. However, as this paper will demonstrate, renewable energy presents significant opportunities for increasing energy access which in turn stimulates economic development, and social inclusion in emerging markets, especially those in tropical regions like Kenya.
|
|||
|
|
|
|||
|
|
\subsection{Hypothesis}
|
|||
|
|
The main hypothesis is as such: Share of renewable electricity positively correlates with the access to electricity in Kenya.
|
|||
|
|
\begin{itemize}
|
|||
|
|
\item{\(H_0\)} There is no positive correlation between share of renewable electricity and access to electricity in Kenya.
|
|||
|
|
\item{\(H_1\)} There is correlation between share of renewable electricity and access to electricity in Kenya.
|
|||
|
|
\end{itemize}
|
|||
|
|
|
|||
|
|
\subsection{Verification criteria}
|
|||
|
|
\begin{itemize}
|
|||
|
|
\item{\bf{R-squared} \(R^2\)} - \(R^2 \ge 0.7\)
|
|||
|
|
\item{\bf{Correlation coefficient} \(r\)} - \(\ r \ge 0.7 \)
|
|||
|
|
\item{\bf{p-value}} - \( \textrm{p-value} \le 0.05 \)
|
|||
|
|
\end{itemize}
|
|||
|
|
|
|||
|
|
\subsubsection{Variables}
|
|||
|
|
\begin{itemize}
|
|||
|
|
\item{\bf{Independent variables}}
|
|||
|
|
\begin{itemize}
|
|||
|
|
\item Access to electricity (percentage of population with access)
|
|||
|
|
\item Year-on-year GDP growth rate in (\%)
|
|||
|
|
\end{itemize}
|
|||
|
|
\item{\bf{Dependent variable}}
|
|||
|
|
\begin{itemize}
|
|||
|
|
\item Share of electricity production from renewable sources (percentage of total electricity generated from renewable sources)
|
|||
|
|
\end{itemize}
|
|||
|
|
\end{itemize}
|
|||
|
|
|
|||
|
|
|
|||
|
|
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
|||
|
|
\newpage
|
|||
|
|
\section{Theoretical part}
|
|||
|
|
As described in Kenneth Lee's chapter of Introduction to Development Engineering , electricity is widely seen as a major driver of economic development and electricity consumption nearly perfectly correlates with GDP per capita \cite{kenya-expanding-electricity-access}.
|
|||
|
|
As access to electricity rises electric lighting extends workdays, therefore increases labor demand. Thanks to innovations in renewable energy, those countries don't have to rely mainly on fossil fuels in electricity generation, which strongly contributes to climate change and its effects on the planet's fragile ecosystem. Developing countries are expected to drive considerable amount of growth in global energy consumption \cite{wolfram2013-energy-demand-developing}.
|
|||
|
|
|
|||
|
|
This poses a a developmental challenge. How can electricity access be expanded in countries with high rate of energy poverty while mitigating the consequences of climate change?
|
|||
|
|
Kenneth Lee discussed the problems of microgrids as a potetntial solution to such problem in rural Kenya with pay-as-you-go business model allowing consumers to buy products on credit. They identified the problem of rural Kenyan infrastructure was not the remoteness of villages, but high cost of installation of low-voltage distribution transformer, which cost around \$398~USD, while the annual per-capita income was bellow \$1000~USD in most rural households. Vast majority of the population without electricity access in Kenya was therefore not off-grid but "under-grid" per se and installing solar microgrids is not a viable long term solution. In their survey of new installation, they did not see significant impacts in the short term with no clear indication as to why. As noted in Khandker et al. about rural electrification in India \cite{khandker-benefits-of-rural-el}, the gains from rural electrification could be much greater for wealthier households which could exacerbate economic inequalities.
|
|||
|
|
|
|||
|
|
In Bernards impact analysis of rural electrification in Sub-Saharan Africa \cite{bernard-rural-electrification}, the author identifies three major time periods in the development of electrification. The first being around the year 1980 when the electrification of rural locations was seen as a key solution to stop migration from rural to urban areas and a source of economic growth.
|
|||
|
|
The second period is the 80s and beginning od 90s, when rural electrification programmes were re-evaluated due to their high costs and disappointing impact. Low connection rates and limited productive use of electricity were observed.
|
|||
|
|
In the third period since the 90s until now, rural electrification is seen as a necessary condition for achieving the Millennium Development Goals and fighting poverty. Programs seek to address the problems of low connection rates and limited productive use through integrated projects and targeted subsidies.
|
|||
|
|
|
|||
|
|
Bernard indentifes the key issues when it comes to low connection rates in rural areas as large upfront cost to connect to the grid, low benefit expectation and the fear of not understanding the billing system.
|
|||
|
|
|
|||
|
|
Other identified problem was the lack of productive use. Electricity is still mainly used for lighting, radios, and televisions in rural areas. Productive use for agriculture, crafts and services is much lower than expected. Bernard propses the reasons for this to be lack of economic opportunities, access to finance and lack of integrated development plans with education on potential usage in mind.
|
|||
|
|
|
|||
|
|
Another problem is the lack of known impacts. The funding for rural electrification programmes is often based on their supposed impacts on health, education or poverty, there is very little empirical evidence to support such claims. Bernard claims that this can be due to, among other things, difficulties in measuring the impacts of infrastructure programmes, long causal chains and attribution problems.
|
|||
|
|
|
|||
|
|
In Khandker et al. article about impacts of rural electrification \cite{khandker-benefits-of-rural-el} they analyzed the data of a large scale survey of households in India and found several key benefits. They found that rural electrification helps youth lower the time spent finding firewood and increases the time spent studying. Increases the labor supply of both men and women, school enrolment, per capita household income and expenditure and helps reduce poverty. However, most of those benefits accure to wealthier rural households, while poorer households use electricity to a limited extent. They identified that limited electricity supply due to frequent outages negatively affects households' connection and consumption and therefore the possible benefits to rural areas. The study concludes, that while rural electrification brings large benefits, it is necessary to stabilize the supply and ensuring that poorer households also benefit.
|
|||
|
|
|
|||
|
|
The optimal strategy for electrification of Kenya was studied by Moner-Girona et al \cite{moner-girona-electrification}. They conducted an extensive spatial mapping of the existing energy infrastructure in Kenya and developed a spatial model of rural electrification called RE\_RU\_KE, taking into account the current state of the energy sector and local resources. The model considers the potential of conventiaonal approaches (diesel generators), clean technologies (solar, wind, hydro, mini-grids), hybrid systems and central grid extension to electrify Kenya at the lowest possible cost.
|
|||
|
|
|
|||
|
|
They conclude, that renewable energy plays a pivotal role in decentralized energy system allowing for energy access in rural areas for competitive prices. Solar power dominance in remote areas which are separated by more than 10km from the grid. According to their modelling, solar generation could make electricity available to 5.98 million people (1284 MW) and hybrid mini-grids could electrify additional 390 thousand people (115 MW). They observe that around 370 thousand people can be covered by diesel generator (33MW). However they conclude that the identified isolated diesel generators are expensive to operate, and maintain and for these reasons. They also reference findings of Lee \cite{kenya-expanding-electricity-access} about the "under-grid" population that those affected are better off investing €150-200 EUR in a stand-alone solar solution as a first phase of electrification overcoming the challenge of connection fees to mini-grids or grid connections. They stress the need to update the Kenya's National Rural Electrification Plan to reflect the decline in price of renewable energy technologies and their increased competititveness aginst diesel generators.
|
|||
|
|
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
|||
|
|
\section{Practical part}
|
|||
|
|
|
|||
|
|
Following table \ref{tab:renewable-electricity-access} is the result of aggregation of several datasources using a python script. Python and it's libraries Pandas, Statsmodels, Matplotlib and Seaborn were also used to perform the data analysis. Source code is available in a public git repository \cite{git-repo}.
|
|||
|
|
|
|||
|
|
\begin{table}[ht!]
|
|||
|
|
\centering
|
|||
|
|
\begin{tabularx}{\textwidth}{p{0.8cm} p{1.7cm} p{2.3cm} p{2.5cm} p{2cm}}
|
|||
|
|
\toprule
|
|||
|
|
Year & GDP per capita (USD) & Access to electricity (\%) & Share of renewable electricity (\%) & GDP growth rate (\%)\\
|
|||
|
|
\midrule
|
|||
|
|
2000 & 617.139 & 15.175 & 40.371 & -4.745 \\
|
|||
|
|
2001 & 617.047 & 17.048 & 59.459 & -0.014 \\
|
|||
|
|
2002 & 611.893 & 18.912 & 67.572 & -0.835 \\
|
|||
|
|
2003 & 668.475 & 16.000 & 73.357 & 9.247 \\
|
|||
|
|
2004 & 692.709 & 22.642 & 62.824 & 3.625 \\
|
|||
|
|
2005 & 778.323 & 24.522 & 59.701 & 12.359 \\
|
|||
|
|
2006 & 854.981 & 26.422 & 56.591 & 9.849 \\
|
|||
|
|
2007 & 1028.226 & 28.342 & 67.016 & 20.263 \\
|
|||
|
|
2008 & 1118.755 & 30.280 & 65.040 & 8.804 \\
|
|||
|
|
2009 & 1123.268 & 23.000 & 53.435 & 0.403 \\
|
|||
|
|
2010 & 1176.311 & 19.200 & 67.877 & 4.722 \\
|
|||
|
|
2011 & 1178.599 & 36.157 & 63.218 & 0.194 \\
|
|||
|
|
2012 & 1396.220 & 38.125 & 71.725 & 18.464 \\
|
|||
|
|
2013 & 1490.422 & 40.092 & 72.524 & 6.746 \\
|
|||
|
|
2014 & 1613.101 & 36.000 & 71.304 & 8.231 \\
|
|||
|
|
2015 & 1625.176 & 41.600 & 85.015 & 0.748 \\
|
|||
|
|
2016 & 1688.852 & 53.100 & 83.884 & 3.918 \\
|
|||
|
|
2017 & 1805.398 & 55.831 & 74.422 & 6.900 \\
|
|||
|
|
2018 & 1987.302 & 61.180 & 85.093 & 10.075 \\
|
|||
|
|
2019 & 2107.735 & 69.700 & 86.919 & 6.060 \\
|
|||
|
|
2020 & 2067.987 & 71.492 & 92.327 & -1.885 \\
|
|||
|
|
2021 & 2208.691 & 76.542 & 90.057 & 6.803 \\
|
|||
|
|
\bottomrule
|
|||
|
|
\end{tabularx}
|
|||
|
|
\caption{Table showing the relationship between electricity access, renewable share, and GDP growth in Kenya. Full data in \cite{git-repo}}
|
|||
|
|
\label{tab:renewable-electricity-access}
|
|||
|
|
|
|||
|
|
\vspace{0.5cm}
|
|||
|
|
\small Sources: World Bank\cite{data-access-to-electricity}, Ember\cite{data-electricity-by-source}, IMF\cite{data-gdp-per-capita}
|
|||
|
|
\end{table}
|
|||
|
|
|
|||
|
|
The following figure \ref{fig:regression} shows significant correlation between Access to electricity and Share of Renewable Electricity in Kenya.
|
|||
|
|
|
|||
|
|
\begin{figure}[ht!]
|
|||
|
|
\includegraphics[width=\textwidth]{regression.png}
|
|||
|
|
\caption{Scatter plot with regression line showing the relationship between electricity access and the share of renewable electricity in Kenya.}
|
|||
|
|
\label{fig:regression}
|
|||
|
|
\end{figure}
|
|||
|
|
|
|||
|
|
\newpage
|
|||
|
|
\begin{itemize}
|
|||
|
|
\item{\bf{R-squared:}} The coefficient of determination for the regression model is \(R^2 = 0.704\), indicating a strong relationship between the variables.
|
|||
|
|
\item{\bf{Correlation Coefficients:}}
|
|||
|
|
\begin{itemize}
|
|||
|
|
\item{\bf{Electricity access and share of renewables:}} \(r = 0.834\), showing a strong positive correlation.
|
|||
|
|
\item{\bf{Electricity access and GDP growth:}} \(r = 0.048\), indicating a weak correlation.
|
|||
|
|
\item{\bf{GDP growth and share of renewables:}} \(r = 0.125\), also a weak correlation.
|
|||
|
|
\end{itemize}
|
|||
|
|
\item{\bf{Electricity Access Coefficient:}} The regression coefficient is 0.561, statistically significant with \(\textrm{p-value} < 0.0001\).
|
|||
|
|
\item{\bf{GDP Growth Coefficient:}} The regression coefficient is 0.177, not statistically significant with \(\textrm{p-value} = 0.505\).
|
|||
|
|
\end{itemize}
|
|||
|
|
|
|||
|
|
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
|||
|
|
\newpage
|
|||
|
|
\section{Conclusion}
|
|||
|
|
The author concludes that there is a strong positive correlation between the share of renewable energy and access to electricity. This finding supports hypothesis \(H_1\) which proposed the existence of such a correlation.
|
|||
|
|
Data analysis revealed that the correlation coefficient between access to electricity and renewable energy share is 0.834, indicating a strong positive correlation. The regression coefficient for access to electricity is 0.561 and is statistically significant with a p-value of less than 0.0001.
|
|||
|
|
|
|||
|
|
The paper further indicates that there is a weak correlation between access to electricity and GDP growth (correlation coefficient of 0.048) and also a weak correlation between GDP growth and the share of renewable energy (correlation coefficient of 0.125). The regression coefficient for GDP growth is 0.177 and is not statistically significant with a p-value of 0.505.
|
|||
|
|
|
|||
|
|
Thus, the conclusion of the paper confirms that the share of renewable energy plays a significant role in expanding access to electricity in Kenya.
|
|||
|
|
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
|||
|
|
% --- Bibliography
|
|||
|
|
\newpage
|
|||
|
|
\printbibliography
|
|||
|
|
\end{document}
|