Image
solar panels for water pumps, awash, ethiopia

Project Details

Country Programme
Efficient & Productive Use
Grid Access
Jörg Peters
Jay Taneja
Gunther Bensch
Hailemariam Teklewold
Ethiopia
RWI Leibniz Institute for Economic Research
Image
Professor Jörg Peters

Jörg Peters

Lead Principal Investigator
Image
Jay Taneja

Jay Taneja

Principal Investigator
Image
Gunther Bensch

Gunther Bensch

Energy and Survey Specialist

Electricity demand forecasting in agriculture

Harvesting the synergies of machine learning and survey data for electrification planning in Ethiopia

Background, challenges and context

While Ethiopia is currently undertaking a National Electrification Strategy, with hydroelectric resources being developed at a high pace, research suggests that rural households, which comprise 80 per cent of Ethiopia’s population, will have very low consumption patterns for quite some time. Therefore, there is likely to be excess supply, representing both enormous opportunities as well as significant challenges.

To make the most of the opportunities and proactively address challenges, planning tools need to be improved, and data-driven projections of electricity consumption need to be developed and used.

Effective planning tools are pivotal in defining the most suitable electrification strategy for different sub-national regions, comparing the extension of the national grid to the establishment of mini-grids or other off-grid systems. These tools have significant potential to improve national electrification masterplans, or provide the necessary basis where such plans do not yet exist. At the same time, there is considerable scope for increasing the accuracy of demand forecasts, which can have a significant impact on location, sizing, and timing of generation and transmission infrastructure.

Demand forecasting models do not yet account for potential productive use hubs, including agricultural demand in specific regions. The agricultural sector is the backbone of large parts of the rural African economy, responsible for a large share of self-employed and employed workers. Fostering electricity consumption for productive purposes, especially in the agricultural sector (and thereby improving revenue collection), will be key for policymakers to assure long-term sustainability of grid access policies.

Understanding, forecasting, and prioritising the varying electricity demand for agricultural and other productive uses is of high importance for Ethiopia. The Ministry of Water, Irrigation, and Electricity (MoWIE) and the Ministry of Agriculture (MoA) have repeatedly sought more research on electrification planning related to productive use, and particularly agriculture.

 

Research overview and objectives 

This research project will inform stakeholders in the agricultural sector on understanding, forecasting, and prioritising electricity demand for agricultural purposes. The esearch questions are:

  • How can a utility develop effective electricity demand forecasts and stimulation techniques for productive uses in agriculture?

  • How can machine learning techniques best be combined with classical on-the-ground surveys to yield informative demand forecasts regarding productive uses in agriculture that ultimately also inform electricity system expansion?

  • Which agricultural demand stimulation interventions related to irrigation and agro-processing can be derived from these demand forecasts?

The team will combine an innovative machine learning prototype simulation model with classical ‘on-the-ground’ surveys among households, enterprises, and communities. The aim is to develop a scalable, cost-effective approach that uses relatively ‘expensive to collect’ information by carrying out surveys in a limited number of regions (a sample size of around 100 villages across Ethiopia will be suitable) and train machine algorithms to extrapolate this in-depth information to the entire country.

This will be done in an iterative way, and will include four steps. The model will first learn basic relationships between satellite imagery and microdata. Secondly, the model will be enhanced using the rich data from the on-the-ground surveys, which will specialise the model for the Ethiopian context. In the third step of extrapolation, satellite imagery will be used, as well as available secondary data. The fourth step comprises ground-truthing, where a subset of the survey data collected will be used to test the efficacy of the approach.

With an understanding of areas with elevated latent electricity demand, the team will work with electricity system planners and analysts from the Electrification Directorate in MoWIE to incorporate the findings from the model into electricity planning processes.

For the region under consideration, they will build a model that can help to determine the cost of electrifying particular locations by grid extension, mini-grid, or dedicated solar PV, depending on the scale and scope of electricity activities in the areas and the current footprint of the electricity grid. The latest state of knowledge on decentralised energy, storage, and smart grids will be taken into account.

The collected and generated data will be used to derive data-driven suggestions for demand stimulation interventions, providing insights on where these activities will be most promising. The team will map the demand for new irrigation and agro-processing potentials, paying attention to the most promising value chains.

The project intends to inform Ethiopia’s ambition to foster agro-industrial growth, and aims to develop approaches that will help to identify potential high-demand regions where the grid roll-out should be concentrated.

 

Local partner

Environment and Climate Research Center (ECRC) at the Policy Studies Institute (PSI)