According to the International Energy Agency (IEA), over 800 million people have no reliable access to electricity, primarily in Sub-Saharan Africa and rural regions of developing Asia (IEA, 2023). However, despite many existing initiatives, energy poverty eradication is a demanding challenge and requires novel strategies to meet the ambitious targets (Hakiri et al., 2016; Longe, 2020; Bacar et al., 2022). Obstacles include high capital costs (Buchana and Ustun, 2015), difficulties with energy demand forecasts (Asiegbu et al., 2023), environmental factors (Kolokotsa and Santamouris, 2014), theft, and damage (Wabukala et al., 2023).

The roll-out of grid or micro-grid connections to rural communities is often limited by the expendable income of a community being too low to purchase sufficient energy to be attractive to an infrastructure investor.

In Malawi, the majority of the rural populations are subsistence farmers arranged in smallholding households. The average annual production of maize per adult in smallholder households is 175.0 kg, which falls short of the minimum subsistence requirement of 200.0 kg. They obtain no external monetary income, and seasonal external work is largely paid in-kind with maize. A culture that would be capable of supporting a self-funding and profitable electrical grid/micro-grid infrastructure does not exist (Eales et al., 2020a).

Malawi is a small country with an estimated land area of 11.8 million hectares, of which Lake Malawi occupies one-fifth of the total. In Malawi, lands with slopes of less than 6.0% account for approximately 63.0% of the total land area, and lands with moderately/strongly sloping terrains with slopes of 6.0–30.0% account for 31.5% (Guiying et al., 2017). Out of 9.4 million hectares of land, approximately 5.3 million ha, or 56.0 percent, was cultivated in 2010. Most of the suitable lands have already been developed for cultivation, and agriculturally available land expansions are limited (Guiying et al., 2017). Malawi has a subtropical climate with distinct rainy and dry seasons: the rainy season spans from November to the following April, delivering 90.0%–95.0% of the annual precipitation, while the mean surface temperature is between 22.0 and 23.0°C (World Bank Group, 2024). Agriculture is pivotal to Malawi’s economy, accounting for over a quarter of its GDP and significantly influencing employment and export earnings. The sector is dominated by maize and tobacco, and as quoted in different publications, maize accounts for over 40.0% of export revenues (Trade, 2024).

There have been some published works looking at micro-grids in Africa and their success or otherwise, including in Malawi (Ikejemba et al., 2017; Eales et al., 2020b; Eales et al., 2022). The conclusions of most of the published works are that these fail due to lack of management following installation and equipment, which is left to deteriorate (Ikejemba et al., 2017). There needs to be sufficient revenue generation for ensuring ongoing management. The costs of hardware for a micro-grid solar and battery installation are approximately US$9.0k–US$10.0k per kW (Eales et al., 2022), while operational costs for ongoing management are quoted as US$3.8k per year. Implementation of solar micro-grids would cost approximately US$210.00 per person and average operational expenses of US$17.00 per person per year (Eales et al., 2020b).

Across large areas of Malawi, a heavily compacted layer of rock-hard earth is found a few inches under the topsoil. This is largely the result of long-term use of manual tilling hoes, human and animal footfall, and occasionally heavy machinery. Plant roots generally cannot penetrate through this hard layer, nor can air or water, which is necessary for the healthy living soils that support good agriculture (Mvula, 2021). The UNDP-quoted figures for soil loss from 2014 estimated that the average national soil loss rate in Malawi was 29.0 tons per hectare per year. However, these are old statistics with no new national statistics available (Vargas and Omuto, 2024). Statistics on the soil loss explains why Malawi is one of the world’s poorest countries, and why hunger is so common.

To generate wealth to pay for electrification, it is advisable to also address agricultural issues, which would help generate rural income to fund electricity. Previous work into joint agriculture/electrification projects has focused on cold storage (Yan et al., 2023) and crop shading (Aroca-Delgado et al., 2018).

Published work on use of mechanical power devices for agriculture in Sub-Saharan Africa includes (Mrema et al., 2018)

• Tractors using traditional two-axle, four-wheel tractors (4 W T).

Some work has been announced on electrification of tractors, but these are likely to be several orders of magnitude more expensive than a Malawian household could afford (Dezeen, 2023).

Deep bed farming (DBF) is an innovative agricultural methodology that has been developed by Tiyeni. DBF reconceptualizes the traditional Malawian agricultural cycle, and the methodology has received approval from the Malawian Ministry of Agriculture. DBF demonstrates a significant increase in yield.

The steps involved with this process are as follows (Tiyeni, 2023):

• Break up the hardpan: breaking up the hardpan delivers immediate benefits, allowing roots, water, and air to penetrate into the soil, curbing soil erosion.

DBF is not just about creating and maintaining the deep bed but also includes a change in other agricultural practices such as soil health and fertility improvement through manure and compost application, residue incorporation, mulching, green manure cover crops, agroforestry, and crop rotation with legumes. This is to encourage farmers to reduce fertilizer use.

DBF requires a substantial increase in the amount of land prepared in the first year before an increase in yield. This creates a barrier to entry, especially for those households with a deficiency in family labor (child-, female-, and elder population-dominant households).

This paper presents a novel solution to these challenges. It proposes a solar-powered electric walk-behind tractor (AfTrak) designed for the preparation and maintenance of fields for the Tiyeni DBF concept, shown in Figure 1. The unit also provides access to zero carbon solar-powered electricity that is reliable year-round. Malawi has only 25.0% variance in solar generation across the year (Madhlopa, 2006). AfTrak directly addresses United Nations Sustainable Development Goal 7, SDG7, i.e., access to affordable and clean energy, on two key fronts:

1. Deep bed farming has been shown to increase smallholder revenue by up to 12 times, potentially generating a self-sustaining model.

This energy could offset high-carbon processes such as charcoal generation, open fire cooking, and lighting.

This paper is novel because it describes and details a new agricultural device, AfTrak, that is capable of both DBF ground preparation and electricity provision. The remainder of the paper is organized as follows: Section 2 highlights the specifications, Section 3 describes the materials involved, Section 4 describes the methods, and Section 5 and Section 6 detail the results and discussions respectively.

The AfTrak system needs to enable faster and less arduous deep bed preparation. It must especially benefit those members of the communities who live in households with reduced resources to conduct manual labor, such as elderly, disabled, child, or female-led households, while at the same time providing a source of affordable electricity to the local community. The requirements for the AfTrak system are split into four parts.

• Performance of the base station (solar charging and energy storage).

This paper focuses primarily on the first two requirements regarding technical performance. Table 1 shows the requirements for the AfTrak unit and the performance requirements for the base station.

To test the system and ensure it worked in line with design calculations, a dawn–dusk test was set up at Loughborough University, as shown in Figure 6. The AfTrak system setup included:

Primary solar array:

• Consists of 12 panels with 425.0-W modules each.

Secondary solar array:

• Two 425.0-W panels mounted on the prototype modular frame.

Battery pack:

• Base station pack made up of eight 12.0 V VARTA 210.0 amp-hour batteries in 2P4S.

Ground station:

• Receives up to 5.0 kW of AC load from the inverter.

AfTrak unit:

• Connects to the base station battery system via a detachable Anderson charging connector.

The setup was designed to operate close to one-third of a full-scale 15.0 kWp system. A 15.0 kWp system was chosen as this is the peak power of solar panels needed to meet a goal of 60 kWh of daily energy supplied as specified by the Milken–Motsepe prize for green energy for their size of micro-grid (Milken-Motsepe prize, 2024).

The measurement system for the test setup is shown in Figure 7. This consisted of commercial AC meters, Fdit, Intelligent Mini Power meter plug (accuracy not given), a Victron Cerbo GX (accuracy not given), and a Victron DC SmartShunt (±0.01 A and ±0.01 V on data sheet) to monitor the power use. The solar radiation was measured using a Kipp & Zonen CMP11 thermopile pyranometer (sensitivity 5.0–32.0 μV/W/m² from data sheet) and Campbell Scientific CR800 logger.

A set of loads as shown in Table 4 were switched on and off during the day on the different devices, and the energy figures on each instrument were captured at the start, middle, and end of the day.

The calculated energy from the solar radiation in conjunction with the solar panel area and energy generation reported by the Victron Cerbo GX is shown in Figure 8; the latter is divided into two parts, the small solar module frame with two panels and a more standard solar array with 12 panels to understand the impact of the frame design on generation. It should be noted that

• All measurements have been taken from an unobstructed location.

The 1 hour shift between calculated energy and measured in time is due to GMT/BST reporting.

The generation as recorded on the two MPPT units over the day was as follows:

• RS 48/6000 MPPT: 29.4 kWh (5,100.0 kWp).

Generated PV energy has exceeded the expected targets for the array. This is due to several factors including slightly longer days at the test location latitude than in the target market, and generally clear sky. The team has optimized the installation using optimal tilt and angle for maximum irradiance; these values can be optimized at any global location, and instructions and a link to a suitable phone app will aid within the country setup. Negative impactors include a high panel temperature (higher than peak seasonal generation in Malawi).

In order to assess the performance of the new modular solar frame, the results were normalized using the total installed capacity, allowing for direct comparison.

The data show that the modular frame consistently generates more energy than the stock frame and over the course of 24 h generated 102.0% as the array.

These improvements have been attributed to two key factors:

• Shadowing: the increased height of the frame stops low level shadowing from grass and insects.

The modular frame design was shown to provide good parking location and a charging station for the AfTrak units, protecting the tractor somewhat from continual thermal radiation, which will extend the units’ lifespan.

The steel framework was demonstrated to be sturdy and will likely hold up to storms, but this will need verifying in the country.

The AfTrak tractor units were successful in charging from the base station array, making use of the solar generation and additional input from the base station batteries. It has been extrapolated that a full charge after ploughing will take 1–3 h dependent on solar radiation.

The charge connector is designed to disconnect the tooling when on charge to prevent accidental operation. This functionality was tested and validated. The system has been tested as an isolated grid. However, our choice of power electronics means that micro-grid operation is possible if required at any stage.

A range of loads were tested through the test period and connected to both the base station and tractor unit; these are shown in Table 4. Two significant base loads were present from an EV charger and water heater, which when combined pushed the continuous load to 4.5 kW with additional 500.0 W of instantaneous power from smaller loads leaving a peak inverter load of 5.0 kW. A total consumption of 34.2 kWh was demonstrated throughout the day. The recorded data during the day are shown in Table 5.

The AfTrak unit used 3.0 kWh of load throughout the day on providing power for devices and running the deep bed farming trenching. This was then recharged using the solar panels toward the end of the day. The battery pack on the base station ended at 2.08 kWh lower charge than it started while the AfTrak battery unit was 0.02 kWh lower. This accounts for inefficiencies within the system.

Although difficult to deconvolve, manufacturing equipment’s efficiency has been stated as

• Victron Inverter at 30.0°C: 96.0%

Efficiency of system = output power/input power = 34.2/(34.2 + 2.08 + 0.02) = 94.2%

A total system efficiency of 94.2% is a reasonable figure considering that a significant amount of energy is directly consumed by the load and not charging the batteries. In a scenario where the batteries are used as short-term storage, an 80.0% efficiency is expected as energy loss occurs at both charging and discharging.

The DBF’s main requirement, from a labor standpoint, is the depth of soil penetration during the tilling process. The AfTrak unit has been demonstrated to be able to successfully mechanize the process of DBF by operating a 400.0-mm-deep cut, while de-compacting the soil to a small particle size, allowing for greater crop root penetration than by use of traditional hand tools, as shown in Figure 9.

The AfTrak tractor is equipped with a smart current monitoring shunt displaying power data live over Bluetooth. During soil preparation, a peak current of 152.99 A was recorded. This will vary with soil conditions but is expected to drop in dryer soils.

The AfTrak digging technique has been measured turning at 400.0 mm for a battery life equivalent to 50.0 m² over 2 h of operation per unit. Combined with the solar base station, each AfTrak unit should be capable of at least two operations per day, resulting in 100.0 m² per unit per day when used with two cycles and 150.0 m² when three cycles are used. The average farm area in Malawi is 0.47 ha (FAO, 2024). To prepare half the land for DBF would require around 2 weeks of an AfTrak unit in year 1, and this is process is significantly faster than use of a pickaxe.

This work is aimed at helping communities in Sub-Saharan Africa to adopt DBF as an agricultural practice to increase crop yield, help with water retention, and reduce soil erosion. This paper has described a micro-electric tractor that has been designed to both undertake DBF and also provide communities with a source of static and mobile micro-grid power.

The work has entailed developing a new base station structure to house the AfTrak unit and provide housing for batteries and the power electronics needed for charging and micro-grid applications. Both the base station and AfTrak unit worked with a variety of loads, and the AfTrak unit has the capability to cut through a hard pan to 40.0 cm.

The AfTrak system is designed to have very little running cost and to be maintained by the community. This results in extremely low running costs for the AfTrak system. The costs of the demonstration system from Tables 2 and 3 are base unit: £9,086.00 or USD$ 11,377.26 (just over 1/3 size) and AfTrak: £7,409.84 or USD$ 9,275.64 (each). The hardware costs for a 5.9 kWp solar array and single 10.0 kWh micro-tractor unit are approximately US$ 20,000.00 combined. This is cheaper than currently quoted typical micro-grid prices of US$ 9.0k–US$ 10.0k per kW but does not include labor, installation, and land costs. Some of the costs of a typical micro-grid include laying cables to houses. This has not been included within the costing as the micro-tractor is portable. We anticipate villages having multiple micro-tractors per base station with sizes being scalable.

Using solar modeling, the amount of energy that could be generated by the base station per year is around 10.0 MWh. This is equivalent to a capital cost of US$ 0.06/kWh. For a full-scale base station this reduces to US$ 0.02/kWh with one tractor and US$ 0.03/kWh with three tractors. The current price of electricity in Malawi is listed as 173.70 K/kWh (Mhone, 2023) equivalent to around US$ 0.10/kWh.

This estimation of the capital cost per unit was based on a quoted 30-year lifespan of the panels. It is expected that the AfTrak units will have a lifespan of at least 10 years, but given that they only need to convert each field once to DBF, they can be retired after this point and used solely as energy vectors.

This work is not complete as detailed testing in Malawi needs to be performed. In addition, technical improvements have been identified. In particular, the original wheel drive, while fine on flat hard ground, did not have enough torque for seamless operation while undertaking trenching. The motor drives had to be replaced, and additional gearing was needed to meet the high torque requirements. It is also necessary to undertake large-scale testing in Malawi within the community to understand more deeply their social and technical requirements and improve the design based on this.