At our latest energy seminar Dr Adrià Junyent Ferré from the Department of Electrical and Electronic Engineering led us through his work on the RENGA project. He has written us this blog post covering the same topics and you can also download a PDF of his slides from the talk.

Access to electricity is a key underpinning factor for many development goals such as food security, manufacturing and access to education, healthcare and water.

As of 2017, about 14% of the global population didn’t have access to electricity and the most deprived areas were in Sub-Saharan Africa and Asia. In some countries, this percentage can be as low as 5% (e.g. South Sudan, 2014) and large differences can be found between urban and rural areas. Urban areas are comparatively easier to electrify due to their high population density and good road infrastructure.

Monolithic energy systems

Conventional electrical power systems rely on large scale generation, transmission grids and distribution networks. Transmission grids are the equivalent to motorways, they transport bulk power across long distances from a few big power plants to large load centres. On the other hand, distribution networks are like smaller roads and streets, and they distribute the power to the end users scattered across smaller distances.

This paradigm was established in the mid-twentieth century when power would be generated in large thermal power plants and it proved to be a good solution at the time. However, it has several drawbacks when used in countries with low rates of electricity access. Firstly, transmission lines are costly to build and to maintain and they are hard to justify in locations where electricity demand is likely to be very low for decades. Secondly, thermal generation may not be the best option if it makes the country dependent on imported fossil fuels.

An agile shift for development

While the old power systems paradigm failed to provide convenient solutions for the electrification, alternative approaches fuelled by the drop in price of photovoltaics (PV), power electronics and battery technologies have had some success over the past 15 years. Small-scale autonomous systems with PV are easy to deploy in remote areas and their costs can be low enough to be a viable solution with little or no subsidies.

These generally take different forms and sizes:

• Solar home systems (SHS) are small, typically single-user. They supply power in the range between 1 W (solar lanterns) and 5 kW.
• Minigrids are community-scale systems with power ratings between few kW and few MW and they combine different forms of generation with batteries (e.g. PV plus stationary diesel generators).

SHS and minigrids have different pros and cons. SHS are owned and operated by individual users and they typically pay for them through micro-credits. The first successful commercial experiences attracted many manufacturers competing to develop the cheapest solution, often at the expense of the durability of their products. This created a bad reputation for SHS and in some cases slowed down their adoption.

The benefits of supplying a larger number of users are well-known in power systems engineering. For example, the maximum instantaneous power required to supply 40 users may be calculated to be just 20 times the maximum power of a single user given the low probability of all users consuming maximum power at the same time. Consequently, minigrids can be more cost-effective to build than an equivalent fleet of SHS, but they require a larger coordinated investment.

The long term viability of SHS and minigrids and how they would evolve to become part of a full-scale power system in the future is still an open question. The characteristics of generation, energy storage, network cabling and protection design in these systems are different than those in large-scale utility networks. For example, some manufacturers have chosen unconventional voltages for cost savings, safety and other considerations (e.g. 48 V DC instead of the typical 400 V three-phase AC). In order to avoid some of these incompatibilities, governments may force minigrids to adhere to national standards for distribution network design. Typically, this leads to higher costs and in some cases taints the viability of this type of projects.

RENGA

This is where the RENGA project comes in(Resilient Electricity Networks for a productive Grid Architecture). Started last May, it is a 30-month research grant that started in May 2018 and supported by the GCRF Energy Networks Call from the EPSRC. The programme is led by Imperial College London with our collaborators in BBOXX Ltd., Meshpower Ltd and the ACE-ESD at the University of Rwanda.

Our work is looking at how we can expand an electrical power system that starts out as isolated SHSs via minigrids and eventually connecting them together and to a wider national power grid.

Our programme encompasses three research themes to devise how this can happen.

Technology for safe and cost-effective interconnection

Minigrids can’t be connected together through simple power lines unless their electrical protections and their control systems are redesigned when this happens. However, power electronics could be used to create fully-controlled links that would be much easier to deploy.

The team from the Department of Electrical and Electronic Engineering at Imperial is being led by myself and we are investigating how these links would be designed considering the need for easy deployment, affordability and very safe and reliable operation.

Over the last 10 or so months we have developed a compact power electronic converter that can be used for this purpose. We are validating its operation in our lab later next month and we plan to connect it to Meshpower’s system by the end of the project.

Design of mini-grids and aggregations of mini-grids

Alongside our work on the technical details of the power system the team led by Professor Jenny Nelson and Dr Iain Staffell is working on the design of the minigrids and how demand and generation can be modelled on these kinds of systems.

In particular they are working on the calibration of solar irradiation models tailored for sub-Saharan Africa along with demand modelling for its inclusion in Professor Nelson’s CLOVER tool. CLOVER is a tool will assist in the sizing of generation and energy storage assets in minigrids considering the evolution of demand over time.

This is not the only aspect of minigrid design that is important. Professor Goran Strbac’s team are adapting their cutting-edge distribution network design tools to be used in the electrification context.

These tools evaluate and optimise the performance of different electrical network designs and their operating strategies in terms of things like supply interruptions seen by customers and running costs. This will be a key area in the design of minigrids and to gain understanding about the value of linking multiple minigrids together for communities and individuals.

Infrastructure planning roadmap

Unfortunately there is no “one-size-fits all” solution to the problem of electrification of rural areas but one of the outputs of the project will be tools that can assist governments in the task of planning the expansion of their electrical power system.

At the moment we are focussing on two, quite different countries, Rwanda and Nepal. While both countries face the challenge of developing their electrical power systems, the geographical distribution of their population and their energy resources are different, which provides a chance to compare how different solutions may work in different parts of the world.

Our local partners

We could not do any of this work properly however without the on-the-ground, local knowledge and expertise of the people working in the community. BBOXX Ltd., Meshpower Ltd and the African Centre for Excellence in Energy for Sustainable Development (ACE-ESD) are all key to the success of the project.

BBOXX and Meshpower allow us to tap into a powerful network of communities in Rwanda and Nepal. They have built successful companies in a very young a growing sector and know so much more than us about the actual difficulties faced once you take kit out of the lab and put it outside in real world conditions.

The ACE-ESD at the University of Rwanda has been a key academic partner, thanks to the wonderful Professor Etienne Ntagwirumugara, providing some much needed technical expertise and hosting two of the PhD students funded by RENGA. They have also helped us through their work on building collaborative research capacity in Eastern and Southern Africa giving us a wealth of contacts and experts to help us achieve our aims.

The future

RENGA still has over a year and a half to go but we already have our eyes on the next steps. By November 2020 we hope to have demonstrated our proposed concept of interconnection technology and developed a set of tools for electrification planning available for anyone to use.

After that we hope that we’ll be able to continue to work with the University of Rwanda and the industrial partners in the project to further develop the technology and look for opportunities to apply the insights learned through the project in other countries with lower rates of electrification.

Dr Adrià Junyent Ferré

Adria graduated from the School of Industrial Engineers of Barcelona (ETSEIB) in 2007 and obtained his PhD in Electrical Engineering from the same university in 2011.

He is a lecturer of the Dept. of Electrial and Electronic Engineering at Imperial College since 2014. His area of expertise is power electronics for voltage-source converter HVDC transmission, wind power generation and new forms of low voltage distribution.

His specific focus is on the design of power electronic converters and how they are controlled to interact with electrical networks.