Hello, you already did chapter 1 and 2 of my MASTERS Thesis already for me. (See attached) So normally you know masters thesis consist of 5 chapters right ??..... But in this case my thesis will be 4 chapters instead of 5 chapters - because thats how the professor wants it ... do you understand ? {I will send you a sample of (old student thesis) how the whole thesis will be structured so you will understand better} so you will know how mine will be.In addition, I will send you the chapters you already did for me so you can continue on the SAME DOCUMENT and the copy of my analysis so far for the methodology and also the simulation results to make work easy for you. I am always at your disposal in case you have any questions. INSTITUTE OF AGRICULTURAL ENGINEERING Agricultural Engineering in the Tropics and Subtropics (440e) Prof. Dr. Joachim Mller Master-Thesis Agricultural Sciences in the Tropics and Subtropics cand. M. Sc. Kilian Blumenthal Assessing the potential use of solar energy supply across the maize value chain in Benin Date of submission: 11.11.2019 2 Abstract Climate change is a human made problem with immense impact on the world population. Sustainable energy solutions are required in order to reduce greenhouse gas emissions, the driving factor behind the climatic change. Solar photovoltaic (PV) systems have advanced the access to electricity worldwide and have shown their potential for decentralized energy supply over the last decades. Among the main benefits, solar energy is a clean, inexhaustible and environment-friendly energy source. Therefore, stand-alone or hybrid solar power systems are a promising solution to replace often used conventional generators or machines powered by fossil fuels. Nowadays, PV systems are successfully used in a variety of small-scale applications. Solar energy systems for agriculture are in an early stage of development. Guiding the implementation of solar energy systems is hence essential for building environmentally friendly agricultural value chains. The framework of the present research is on exploring the use of off-grid solutions to drive different technologies along the maize value chain, with a strong focus on which mechanization steps can be powered totally or partially with solar energy. The study bases on a data collection performed in Benin, where interviews with farmers were conducted, power demand of machines measured and information of different processes along the value chain gained. The data was analyzed to draw a conclusion about the energy balance and greenhouse gas emissions along the value chain and to identify suitable processes for the integration of solar energy. Findings show that highest energy input and greenhouse gas emissions lie within the processes of fertilization and milling. Diesel-driven maize mills were identified of having the highest potential for the application of solar energy. Two systems based on solar energy were designed to substitute the diesel motor of current milling systems. Simulations are used to evaluate the designed systems in their performance. The results of a comparison show that the designed solar battery system as well as the hybrid PV-diesel system have both, financial and environmental advantages towards the purely diesel system. This study presents an approach of assessing the potential of solar energy across the maize value chain. However, the approach also allows the analysis of other value chains and can hence guide researchers, policymakers and end users on how to best power mechanization with solar energy. 3 Table of content 1 Introduction ......................................................................................................................11 1.1 Study area ..............................................................................................................11 1.2 Maize production...................................................................................................14 1.3 Maize value chain..................................................................................................14 1.4 Energy use and greenhouse gas emissions ............................................................15 1.5 Solar energy systems.............................................................................................16 1.5.1 Photovoltaic system components......................................................................17 1.5.1.1 Photovoltaic PV modules.............................................................................17 1.5.1.2 Batteries........................................................................................................19 1.5.1.3 Charge Controller.........................................................................................20 1.5.1.4 Inverter.........................................................................................................20 1.5.2 PV-diesel hybrid systems..................................................................................21 1.6 Aim of the study ....................................................................................................23 2 Material and Methods.......................................................................................................24 2.1 Study Area .............................................................................................................24 2.2 Measurement devices.............................................................................................25 2.2.1 Weather monitoring ..........................................................................................25 2.2.2 Fuel consumption measurements......................................................................26 2.2.3 Weight measurements.......................................................................................27 2.2.4 Power consumption measurements...................................................................27 2.2.5 Global positioning system (GPS) data logger...................................................28 2.2.6 Sensor equipment summary..............................................................................28 2.3 Methods .................................................................................................................29 2.3.1 Weather data .....................................................................................................29 2.3.2 GPS interpretation.............................................................................................29 2.3.3 Questionnaire-based survey..............................................................................29 2.3.4 Fuel consumption measurements......................................................................30 2.3.5 Electricity consumption measurements ............................................................32 2.3.6 Energy Analysis................................................................................................33 2.3.6.1 Energy and GHG emission conversion factors............................................34 4 2.3.6.2 Energy use efficiency...................................................................................35 2.3.7 Designing and dimensioning of hybrid and PV system components................35 2.3.8 Financial calculations........................................................................................38 3 Results and discussion......................................................................................................40 3.1 Weather data ..........................................................................................................40 3.2 Farm classification.................................................................................................41 3.3 Maize production chain .........................................................................................41 3.3.1 Preparation ........................................................................................................42 3.3.1.1 Clearing........................................................................................................42 3.3.1.2 Primary and secondary tillage ......................................................................42 3.3.2 Plant production ................................................................................................45 3.3.3 Postharvest ........................................................................................................47 3.3.3.1 Storage..........................................................................................................48 3.3.3.2 Peeling..........................................................................................................48 3.3.3.3 Shelling.........................................................................................................48 3.3.3.4 Milling..........................................................................................................49 3.4 Cropping calendar..................................................................................................51 3.5 Energy input-output analysis.................................................................................52 3.6 Greenhouse gas emissions .....................................................................................55 3.7 System design and simulation ...............................................................................57 3.7.1 Selection of the target processing step..............................................................57 3.7.2 Profile load estimation of the maize mill..........................................................58 3.7.3 Intervention set up.............................................................................................59 3.7.3.1 Baseline (Diesel system)..............................................................................59 3.7.3.2 Intervention 1 (Generator system)................................................................59 3.7.3.3 Intervention 2 (PV system) ..........................................................................59 3.7.3.4 Intervention 3 (PV-diesel hybrid system) ....................................................62 3.7.4 Financial comparison ........................................................................................65 3.7.5 GHG emission comparison ...............................................................................68 4 Conclusion........................................................................................................................70 5 References ........................................................................................................................72 6 Annex................................................................................................................................79 5 Symbols AF days Days of autonomy C USD Initial costs CN Ah Battery Capacity CPW USD Costs present worth d % Discount rate E0 W/m Irradiance at standard test conditions EPW USD Fuel costs present worth G kWh/m Global irradiation I A Current i % Interest rate MPW USD Operation and maintenance costs present worth U V Voltage P W Power Ppeak Wp Peak power RPW USD Replacement costs present worth SPW USD Salvage value present worth Wd kWh/day Daily energy requirement 6 Abbreviations AC Alternating current CC Cycle charging CO2 Carbon dioxide CH4 Methane DC Direct current DOD Depth of discharge EUE Energy use efficiency F-gases Fluorinated gases GPS Global positioning system GHG Greenhouse gas INRAB National Agricultural Research Institute of Benin (Institut National des Recherches Agricoles du Bnin) K Potassium LF Load following LCC Life cycle costs MPP Maximum power point N Nitrogen N2O Nitrous oxide P Phosphorus PR Performance ratio PTAA Agriculture and Food Technology Program (Programme de Technologie Agricole y Alimentaire) PV Photovoltaic PW Present worth SE Specific energy SOC State of charge SSA Sub Saharan Africa STC Standard test conditions USD United States Dollar 7 List of figures Figure 1: The location of Benin ..............................................................................................12 Figure 2: Long term average of global horizontal irradiation (top left) from 1999-2015 (SOLARGIS, 2019), rainfall in mm (top right) in the year 2017 (eAtlas, 2019) in Benin and distribution of precipitation (bottom) during the year in Porto-Novo in the southern region of Benin (Meteonorm yearly averaged data from 1991-2010)..................................................13 Figure 3: Maize yield (left), the cultivated area of maize (middle) and maize production (right) in Benin (eAtlas, 2019).............................................................................................................14 Figure 4: Steps in the global maize value chain (Daly et al., 2017)........................................15 Figure 5: Inflation adjusted module price in /Wp (Fraunhofer ISE, 2019)...........................17 Figure 6: Influence of irradiation (top) and the cell temperature (middle) on the current and voltage characteristics of a solar cell and the characteristics of current-voltage and powervoltage at constant irradiation and temperature and the corresponding MPP (bottom) (Quaschning, 2005) ..................................................................................................................18 Figure 7: Depth of discharge and corresponding expected cycle life of a lead-acid battery (Zhang et al., 2017)...................................................................................................................20 Figure 8: Inverter efficiency of a PV system (Silvestre, 2018)...............................................21 Figure 9: Energy flow and components of a PV-diesel hybrid system (Fodhil et al., 2019) ..22 Figure 10: The sample areas in Adjohoun and Ktou .............................................................24 Figure 11: Satellite image of the farm and the area of maize production (red circle).............25 Figure 12: Installed weather station at the PTAA ...................................................................26 Figure 13: Filling flasks with diesel for measurements of fuel...............................................26 Figure 14: Weighing set up during the assessment .................................................................27 Figure 15: Digital wattmeter for 3 phase powered machines (left) and the AC/DC Mini current clamp (right).............................................................................................................................28 Figure 16: GPS tracker for leveling of the tractor...................................................................28 Figure 17: Measuring fuel volume before filling the tank (left) and wooden stick used to control fuel volume (right)....................................................................................................................30 Figure 18: The tractors used in Adjohoun (left) and Ktou (right) during field preparation ..30 Figure 19: Fuel measurement of the gasoline motor before and after operation (left) and shelling machine in operation (right)........................................................................................31 Figure 20: Milling process for maize grains at the PTAA. .....................................................32 Figure 21: Digital wattmeter connected to the electric motor.................................................33 Figure 22: Irradiation and temperature profile during a typical week in May (top) and during a very cloudy and rainy week in June (bottom) ..........................................................................40 Figure 23: Flow diagram of maize production in Benin .........................................................42 Figure 24: GPS tracking of the different farm operations mulching (left), plowing (middle), secondary tillage by rotavator (right) .......................................................................................43 Figure 25: GPS tracking of the plowing on the field in Ktou................................................44 Figure 26: Traditional silo outside view (left) and maize ears inside the silo (right) .............47 Figure 27: Process of maize peeling........................................................................................48 Figure 28: Typical local maize mill in Porto-Novo.................................................................51 8 Figure 29: Cropping calendar in Benin ...................................................................................52 Figure 30: Daily working hours in four visited maize mills ...................................................58 Figure 31: Hourly profile for the milling machine..................................................................58 Figure 32: Battery SOC, power demand of the load and power generated by the PV generator during 4 typical days.................................................................................................................60 Figure 33: Development of the battery SOC during four days of operation with very low solar irradiation (top) and when the milling machine is not used for one day ..................................61 Figure 34: Life cycle costs (top) for the 20-year period, savings compared to the baseline within 20 years (middle) and payback period compared to the baseline (bottom) of the hybrid system as a function of the share of solar PV.......................................................................................63 Figure 35: Battery SOC, power demand of the load and power generated by the PV generator during a day with high solar irradiation....................................................................................64 Figure 36: Hybrid PV-diesel system behavior under unfavorable conditions ........................65 Figure 37: Financial comparison of the accumulated yearly costs of the baseline and the three interventions over a 20-year period..........................................................................................67 Figure 38: Yearly CO2 mitigation of the hybrid system for different PV shares....................68 Figure 39: Yearly GHG emissions from baseline and interventions.......................................69 9 List of tables Table 1: Summary of the equipment used during experiments...............................................29 Table 2: Data of the tractors used on the farms in Adjohoun and Ktou ................................31 Table 3: Specifications of the diesel engine at PTAA.............................................................31 Table 4: Specifications of the three-phase induction motor at PTAA.....................................32 Table 5: Overview of the data that has been collected and is used in the energy analysis......33 Table 6: Energy equivalents of agricultural inputs..................................................................34 Table 7: Greenhouse gas emissions in CO2 equivalents of agricultural inputs.......................34 Table 8: Overview of the system designs................................................................................36 Table 9: Energy prices for electricity and fuel in Benin..........................................................39 Table 10: Prices for main system components........................................................................39 Table 11: Daily global horizontal irradiation average per month in Porto-Novo in kWh/m (Meteonorm).............................................................................................................................41 Table 12: Classification of total field size and maize field size of the interviewed farmers...41 Table 13: Average values of work input in the land preparation per ha of maize cultivation.42 Table 14: Fuel consumption, time and trajectory of mechanized field operations in Adjohoun ..................................................................................................................................................43 Table 15: Fuel consumption, time and trajectory of mechanized field operation in Ktou ....44 Table 16: Overview of the mechanized field preparation scaled to 1 ha.................................44 Table 17: Agricultural inputs and yield, given as mean value with standard deviation ..........46 Table 18: Average values of work inputs in the land preparation and plant production per ha of maize cultivation.......................................................................................................................47 Table 19: Fuel measurements with the sheller for 122.41 kg of maize ears with husk...........49 Table 20: Time and fuel consumption for the milling of 10 kg of maize in the PTAA mill driven with diesel engine .....................................................................................................................49 Table 21: Time and energy consumption for the milling of 10 kg of maize in the PTAA mill driven with an electric engine...................................................................................................50 Table 22: Time and fuel consumption for the milling of 10 kg of maize in the local maize mill driven by a diesel engine ..........................................................................................................50 Table 23: Overview of the energy input into the postharvest processes scaled to 1000 kg, given are mean value with standard deviation ...................................................................................51 Table 24: Energy input [MJ/ha] along the value chain for farms with high and low mechanization level..................................................................................................................53 Table 25: The energy - input-output, specific energy and energy use efficiency for farms with high mechanization (HM) and low mechanization (LM) level ................................................54 Table 26: Greenhouse gas emissions (kg CO2 eq. / ha) along the value chain for farms with high and low mechanization level ............................................................................................55 Table 27: GHG emissions for high-level mechanized farms (HM) and low-level mechanized farms (LM) for activities until harvest and activities including postharvest............................56 Table 28: Mechanized processing step across the maize value chain in Benin.......................57 Table 29: Investment costs and lifespan of system components.............................................66 10 11 1 Introduction Climate change is one of the mayor global challenges that humankind currently faces. It goes along with global warming, increasing greenhouse gas emissions, environmental pollution and the depletion of natural resources. As most emissions derive from energy systems based on the combustion of fossil fuels, the transition of these systems to a high share of renewable energies is required to allow a sustainable global future (Hake et al., 2015). In the majority of developing countries, agriculture is the primary driver of economic and social development where agricultural production of foods and non-food products and services contribute to food security and poverty reduction. Energy is required along the agricultural production and transformation and is hence a key factor in improving livelihoods and strengthening local development. The access to clean, reliable and affordable energy services at household level but also in agricultural production and processing is required to drive the development on a sustainable basis. A high dependency on fossil fuels makes agricultural production sensitive for the rise of crude oil prices, which would increase production costs and hence, food prices (Utz, 2011). In 2014, about 15% of the global population (1.06 billion people) did not have access to electricity. This deficit is mainly concentrated in Sub-Saharan Africa, where the share of population without access is 62.5%. There are two strategies to increase electricity access: (i) grid electrification to connect urban, peri-urban and rural areas and (ii) off-grid electrification by the implementation of micro- or mini-grid systems on community-level or isolated system on household level. Grid-based electrification is, especially in rural areas, often very challenging. The distribution infrastructure is weak, the costs to connect remote areas high and households with low income are not able to pay the high connection charges. Even with grid connection, the service can be very unreliable with inconsistent supply and frequent power outages. Off-grid solutions can offer reliable alternatives for the electrification, especially in remote areas (The World Bank, 2017). 1.1 Study area Benin is situated in the western part of Africa, as seen in Figure 1, with borders to Togo, Burkina Faso, Niger and Nigeria. The population size is 11.4 million (2018), with more than half of the population (52.7%) living in rural areas. 17.2% of the rural population and 72.5% of the urban population have access to electricity (The World Bank, 2019). The poor electrification rate in rural areas is due to high investment costs for the connection of these areas to the national grid. In Benin, 75-95% of the total electricity supply is imported by neighboring countries, especially Nigeria, Ghana and Togo (Odou et al., 2019). The economy of Benin depends heavily on agriculture, with over 70% of its population engaged in it, mostly in the form of smallholder systems (Amegnaglo and Yao Soglo, 2019), which are estimated to be 550,000, averaging 1.7 hectares (IFAD, 2019). 12 Figure 1: The location of Benin 1 Figure 2 shows the climatic situation in Benin. The yearly average of daily irradiation in the country ranges from 4.6 kWh/m in the southern parts of Benin, to 5.8 kWh/m in the northern parts of Benin. The rainfall in the year 2017 ranged from 637 to 1400 mm, with high rainfalls commonly occurring in the central parts of the country. In the northern parts of the country, there is one long rainy season from May to September, followed by a long dry period. In the southern parts of Benin, there are two distinct rainy seasons. The long rainy season ranges from March to July. The second and short rainy season ranges from September to October. 1 www.maps.google.com 13 Figure 2: Long term average of global horizontal irradiation (top left) from 1999-2015 (SOLARGIS, 2019), rainfall in mm (top right) in the year 2017 (eAtlas, 2019) in Benin and distribution of precipitation (bottom) during the year in Porto-Novo in the southern region of Benin (Meteonorm yearly averaged data from 1991-2010) 0 50 100 150 200 250 300 Monthly precipitation (mm) 14 1.2 Maize production Worldwide, maize (Zea mays l.) is the most widely cultivated crop with a production of over a billion tons per year and is therefore of very high importance for global food security and livelihoods, especially in developing countries. According to (Leroux et al., 2019), maize is the leading staple food in West African countries with a yearly consumption of more than 30 kg/capita. From 2007-2016, yields of rainfed maize systems in Sub Saharan Africa (SSA) range between 1.68 to 1.99 t/ha, which is 15-25% of the yield potential for water-limited production (ten Berge et al., 2019). Maize can be prepared in different forms which include whole-maize foods, porridges or beverages (Ekpa et al., 2018). In Benin, maize is the most important staple food and the most cultivated crop, with 85% of farmers growing it. The area devoted to maize production is about one-third of the total agricultural area in Benin (Amegnaglo, 2018). Figure 3 shows the distribution of yield, cultivated area and production of maize in Benin for the period 2017-2018. The central western part of Benin has the highest productivity with up to 2.7 t/ha, while in the municipality of Ktou it lies between 1.2 and 1.4 t/ha. The most significant area under maize cultivation countrywide is in the municipality of Ktou. The large area devoted to maize production in this municipality makes it the mayor producer of maize in the southern part of Benin. Figure 3: Maize yield (left), the cultivated area of maize (middle) and maize production (right) in Benin (eAtlas, 2019) 1.3 Maize value chain The central actors along the global maize value chain are the input suppliers, farmers, traders or aggregators, processors (mills), food manufacturers, breweries and producers of animal food (Figure 4). Maize production by smallholder farmers or commercial farms requires different inputs, of which most important are land, labor, seeds, water, fertilizers, agrochemicals and 15 farming equipment. After cropping and harvesting, maize can be stored without shelling or shelled and traded for further processing (Daly et al., 2017). According to Daly et al. (2017), the processing of shelled maize into maize flour starts with cleaning, drying and grading. Maize kernels are then grinded. There are several outputs that can be identified depending to the end-use, which can be classified into products for human consumption, production of fuels or animal feed. In West African countries, maize is usually consumed in the form of a cooked paste. Maize is therefore processed to maize flour by local millers, which exist throughout African cities, villages and markets (Kaminski, 2013). Although maize plays a vital role in the nutrition and economy in West Africa, the value added to maize products in these countries is not fully maximized, due to several constraints such as the limited or reduced access to inputs (fertilizers, improved seeds and agrochemicals), adverse weather conditions driven by climate change and lack of labor force, which leads to low productivity. Under this scenario, producers and processors confront high production costs, which mainly are the results of high energy costs. Several challenges also take place at market level. Where farmers do not have access to the prices or only sell small quantities without being organized into cooperatives, speculative traders and agents can dominate the maize value chain and reduce farm gate prices (Ba, 2017). Figure 4: Steps in the global maize value chain (Daly et al., 2017) 1.4 Energy use and greenhouse gas emissions Agricultural systems produce and consume energy. The energy that goes into this sector can be in the form of locally available, non-commercial energy or in the form of direct and indirect commercially available energy (Hamzei and Seyyedi, 2016). According to Saad et al. (2016), direct sources of energy release their energy during application and can be in form of fuel, 16 electricity or animate power (human or animal). Indirect sources of energy are such forms of energy that do not release energy directly but consider the energy input for their production or transport, such as seeds, fertilizers, pesticides or abrasion of machines. Besides the energy input, also GHG emissions can be of direct or indirect form. Greenhouse gas emissions are the main driver for climate change and cluster different gases emitted by human activities, mainly carbon dioxide (CO2), methane (CH4), nitrous oxide (N2O) and fluorinated gases (F-gases). The area of agriculture, forestry and other land uses accounted with 24% of global GHG emissions in 2010, mostly deriving from cultivation of crops, livestock and deforestation. The sector of industry accounted for 21% of total global emissions, mainly due to the combustion of fossil fuels (EPA, 2019). The processes in agriculture that release most emissions are the enteric fermentation in the digestive system of ruminants, the usage of manure, synthetic fertilizers and rice cultivation (Tubiello et al., 2014). Climate change has a significant impact on agriculture, mainly due to the increase of mean temperatures, changes in rainfall patterns, the frequency of extreme weather events such as droughts or floods, salinization and the rise of the sea level (FAO, 2013). Yield reductions are expected, especially in Southern Africa, which could be as high as 30% by 2030 for maize (Lobell et al., 2008). In Benin, the effects of climate change are characterized by a large rainfall deficit, extreme rainfalls and increased temperatures, which resulted in a decrease of yield in recent years. Farmers reported increases in temperature and a decrease of rainfalls, which lead to drought events during cropping. An adaptation of the cropping calendar or the use of improved maize varieties are applied strategies to cope with the effects of climate change (Yao Soglo and Nonvide, 2019). The combustion of fossil fuels for transportation, electricity or heat production is the most significant cross-sectoral source of GHG emissions (EPA, 2019), which makes it indispensable to adapt renewable sources for clean energy production. Solar energy has great potential for replacing partially or totally fossil fuels. It can be used in its direct form using the solar radiation with technologies such as solar thermal power plants or photovoltaics for electricity generation. Indirect forms of solar energy are waterpower, biomass or wind energy, which can be transformed into useful energy with different technologies (Quaschning, 2005). 1.5 Solar energy systems Nowadays, the prices for photovoltaic modules are continually decreasing due to technological improvements and economies of scale. The experience curve shown in Figure 5 illustrates the price development. Which each doubling of the cumulative production, the price decreases by around 25% (Fraunhofer ISE, 2019). Costs are expected to further decrease with increasing production volume, more automation during the production or newer solar cell materials (Quaschning, 2005). The following section will provide information of two options to cover the power demand either totally with photovoltaic systems or partially with hybrid systems. 17 Figure 5: Inflation adjusted module price in /Wp (Fraunhofer ISE, 2019) 1.5.1 Photovoltaic system components 1.5.1.1 Photovoltaic PV modules The global irradiation of sunlight on a horizontal plane consists of a direct and a diffuse component. The direct solar radiation crosses the atmosphere without obstacles such as clouds, which leads to the casting of shadows. Diffuse radiation has no defined direction, as the sunlight is weakened and deflected by clouds, air pollution or fog particles, and therefore does not cast a shadow. Photovoltaic is the most versatile of renewable energy technologies, as it is very modular and scalable and allows a wide range of generator sizes. Solar cells transform the solar energy of the global horizontal irradiation into electrical energy, basing on the photoelectric effect. They usually consist of silicon and can be classified into three main groups: monocrystalline, polycrystalline and amorphous. Cells made out of crystalline materials can have efficiencies of up to 20%, while cells of monocrystalline silicon perform better than polycrystalline, but are more costly to produce. Cells from amorphous silicon have lower production costs, but also lower efficiencies of around 6-8% (Quaschning, 2005). According to Seltmann (2013) the electrical power (P) generated by the solar panel can be calculated from the voltage (U) and the current (I). The voltage of the PV module depends on the cell temperature, whereas the current depends on the irradiation. Figure 6 shows characteristics of solar cells in their dependency on irradiation and temperature as well as their generated power. With an increase in irradiation, also the cell current increases. With increasing temperature, the voltage decreases. The output power of PV is the product of voltage and 18 current. With changing temperature and irradiation, there are different combinations of voltage and current. Each combination has a point where the product of voltage and current is maximum. This point is called Maximum Power Point (MPP). Figure 6: Influence of irradiation (top) and the cell temperature (middle) on the current and voltage characteristics of a solar cell and the characteristics of current-voltage and powervoltage at constant irradiation and temperature and the corresponding MPP (bottom) (Quaschning, 2005) 19 1.5.1.2 Batteries Batteries are storage components for electrical energy, which are charged while adding and discharged while removing an electrical current over time. They consist of electrochemical cells, in which the electrical energy can be stored and released by a reversible conversion of electrode material. The electrode material changes with the type of battery, which are in most cases, for economic reasons, lead-acid batteries. Batteries are used in photovoltaic systems for different reasons. They act as a buffer, to reduce the mismatch between power generated by the PV system and the power requirement of the connected consumer. The power generated depends on the irradiation and thereby varies during the day, while the power consumption of the consumer often requires constant power during a particular time. The power generated by the PV system that surpasses the demand is stored in the batteries and can be released when the power demand is higher than the power generation. Batteries also can provide an operational autonomy, which allows to cover the demand as well on cloudy days or in the case that some components of the PV system fail (Spiers, 2018). The capacity of the battery is given in ampere-hour (Ah), lead-acid batteries have a charge efficiency of around 95%, which means that the Ah discharged are 5% lower than the Ah required for complete charge. Depth of discharge (DOD) and state of charge (SOC) are two parameters that describe the charge of the battery. The DOD is the percentage of the capacity that was taken from the completely charged battery. The SOC describes the percentage of the capacity available in the battery, compared to the fully charged state. (Spiers, 2018) The recommended maximum DOD of a lead-acid battery is 80 % for autonomy reserves, in order to avoid problems deriving from deep discharge. However, when it comes to the regular operation, depths of discharge of higher than 50% should be avoided to allow a longer battery lifetime (Quaschning, 2005). Batteries are often the most expensive components in a PV-battery system. Their correct sizing is hence crucial for the financial viability of the installation. An optimal size has to be found, which does not oversize the batteries for economic reasons, but which also does not undersize the battery capacity, to allow a required autonomy and to avoid deep discharge of the battery (Spiers, 2018). According to (Spiers, 2018), the lifetime of well-maintained batteries in a PV system with constant charging and discharging is limited by the cycle life. A cycle is defined as a discharge followed by a recharge. The cycle life of batteries describes the number of discharge cycles to a specified DOD that the battery can perform until the available capacity decreases to a certain threshold, often 80% of the initial capacity. With higher DOD and temperatures, the cycle life of the battery is reduced. Figure 7 shows the linkage between DOD and expected cycle life of a lead-acid battery. 20 Figure 7: Depth of discharge and corresponding expected cycle life of a lead-acid battery (Zhang et al., 2017) 1.5.1.3 Charge Controller Whenever batteries are used in a PV system, a charge controller is needed to protect them from overcharging or deep discharge, as this would reduce the battery lifetime. The PV modules, but also the batteries and the consumer are connected to the charge controller. Charge controllers usually control the battery voltage. Once a specific battery voltage is reached, the charging is stopped. When the battery is being discharged and the battery voltage falls under a threshold, the consumer is disconnected from the battery (Quaschning, 2005). 1.5.1.4 Inverter As the PV generator and the batteries work on direct current, an inverter is needed in the system if the consumer works on alternating current. The inverter transforms the DC current to AC current. The conversion comes with efficiency losses, which depend on the output power of the inverter, as seen in Figure 8. The conversion efficiency is higher with the output power closer to the power rating of the inverter and can reach a range of 90 95% (Wagner, 2015). 21 Figure 8: Inverter efficiency of a PV system (Silvestre, 2018) 1.5.2 PV-diesel hybrid systems Hybrid renewable energy systems are a combination of at least one renewable energy source, such as photovoltaic or wind generators and one conventional energy source, such as a diesel generator (Fodhil et al., 2019). Widespread types of hybrid systems consist of PV and diesel generators, as they have very complementary characteristics. The capital cost required for PV systems is higher than for diesel systems, as more advanced components are required. Contrary, the operational costs and maintenance costs are lower in the PV systems, as the energy is generated from solar radiation and components require fewer maintenance. Diesel systems can operate at all times, provided that fuel is available, while PV systems depend on the amount of solar radiation (Shaahid and Elhadidy, 2003). Integration of batteries is usually done to reduce the occurring mismatch between power demand and power generation or to cover peak loads that surpass the renewable energy source (Yamegueu et al., 2011). A hybrid system consisting of PV generator, diesel generator, batteries and consumer (also called load) is shown in Figure 9, the arrows represent the corresponding energy flows. Solar panels provide the power to cover the load demand. Surplus energy is used to load the batteries, until they are completely charged. When the load is higher than the power supplied by the PV panels, the missing energy is provided by the batteries based on their state of charge. When the batteries have reached a certain depth of discharge, the diesel generator is used as a backup device to cover the energy demand (Fodhil et al., 2019). The hybrid system contains two types of current: alternating current (AC) of the load and the diesel generator and direct current (DC) of the PV and the batteries. An inverter is hence required for the system to allow a flow of energy between the AC and DC components (Adaramola et al., 2014). Purely renewable energy systems for larger loads require a certain amount of storage in form of batteries to allow a continuous power supply, which involves high capital costs. Diesel 22 systems have high operational costs, especially if the fuel is expensive or the system situated in a rural area. Hybrid renewable energy systems of PV and diesel can reduce the required battery storage and fuel consumption and thereby supply larger loads relative to purely PV systems with significantly reduced costs (Hazelton et al., 2014). In a hybrid system, the diesel generator combusts the diesel fuel and converts it into electricity. The typical fuel efficiency of diesel generators is around 0.33 l/kWh when operated at about 80% of its rated power. When the generator runs at a lower level than his nominated power, the fuel efficiency decreases, hence the fuel consumption per generated kWh increases (Yamegueu et al., 2011). Figure 9: Energy flow and components of a PV-diesel hybrid system (Fodhil et al., 2019) PV-diesel hybrid systems can have different control strategies, which are also called dispatch strategies. Depending on these strategies, the operation mode of the different system components, mainly diesel generator and batteries, is controlled. The dispatch strategy determines the start and stop of the diesel generator and its operation mode as well as the charging and discharging of the batteries and the sources for their charging (Ashari and Nayar, 1999). Of different strategies, the Load Following operation (LF) is often utilized, in which the load is covered by PV and batteries and only in the case of their unavailability, the diesel generator supplies the load. In this case, the batteries are charged only with excess PV, which reduces the operational costs. Another strategy can be to simultaneously supply the load and charge the batteries with the diesel generator (Cycle Charging) (Das and Zaman, 2018). The optimal dispatch strategy and system design are interdependent and must be developed together. Due to the interdependency, the dispatch strategy also depends on the component costs and the renewable energy penetration (Barley and Winn, 1996). 23 1.6 Aim of the study This study aims to assess the potential implementation of solar energy systems to be integrated along the maize value chain in Benin. The specific objectives were as follows: 1. To assess the processes across the maize value chain in Benin; 2. To assess the energy input and greenhouse gas emission output along the maize value chain; 3. To identify processes suitable for an intervention with solar energy; 4. To design technical interventions and to simulate their performance; 5. To evaluate the interventions financially and environmentally. 24 2 Material and Methods The presented work was carried out within the project Program of Accompanying Research for Agricultural Innovation (PARI) as part of the One World, No Hunger Initiative (SEWOH) of the German Ministry for Economic Cooperation and Development (BMZ, grant number 2014.0690.9) in collaboration with the colleagues at the National Agricultural Research Institute of Benin (INRAB) in Porto-Novo Benin. 2.1 Study Area The field research was performed in March and Mai 2019. Part of the data collection was done at the Agriculture and Food Technology Program (PTAA) of INRAB located in Porto-Novo (6.47794 N 2.64061 E). Local partners from INRAB pre-selected mechanized farms in the municipalities Adjohoun (Department Oum) and Ktou (Department Plateau). The farms assessed in Benin are shown in Figure 10. At the same time, farmers were surveyed around the pre-selected farms in each municipality by using a structured questionnaire, to assess the different processes and the energy requirements within the maize value chain. Figure 10: The sample areas in Adjohoun and Ktou2 2 https://drive.google.com/open?id=1JB1amUoRbfCsayGYY9zDgDXHxWrZiK6z&usp=sharing 25 The pre-selected farm in Adjohoun is located at 6.74047 N 2.52836 E. The farmer owns a total area of 77 hectares, large proportion of the farm is used for palm oil production. Other cropping activities in the pre-selected farm are the production of oranges, maize for seed production, rabbit-keeping and fish farming. On this farm, maize is cultivated during the short rainy period, the seeds are then sold for the planting in the long rainy season. At the farm it was found a wide variety of machines (tractors, shelling machine, water pump) as well as attachments for soil and land preparation (mulcher, plow, rotavator). During the field research, the production processes of 0.665 ha of maize were assessed. Figure 11 shows a satellite view of the farm. Figure 11: Satellite image of the farm and the area of maize production (red circle)3 The farm in Ktou is located at 7.32081 N 2.57236 E. The visited farmer owns 40 ha of land, the main crops are maize in rotation with cotton. The area assessed in this farm has 1.2 ha. The farmer owns several machines (tractor, trailer) and attachments for soil preparation (plow, disc harrow). 2.2 Measurement devices 2.2.1 Weather monitoring A weather station Ambient Weather WS-1002-WIFI OBSERVER was installed on the compound of the PTAA. Its sensor was fixed at the height of 6 meters on a metal pole, oriented in southern direction and surpassing the rooftop to avoid shading, as seen in Figure 12. 3 https://drive.google.com/open?id=1JB1amUoRbfCsayGYY9zDgDXHxWrZiK6z&usp=sharing 26 Figure 12: Installed weather station at the PTAA The distance of the weather station at PTAA to the fields of the selected farmers is 30 km (distance to Adjohoun) and 100 km (distance to Ktou). Additional irradiation data for Porto-Novo was obtained by the software Meteonorm, which bases on the average of the years between 1991 - 2010. The Meteonorm data for the location of the PTAA is a result of interpolation of 3 weather stations: Cotonou (32 km distance), Lom airport (158 km) and Douala (832 km). Meteonorm provides year-around data for irradiation on an hourly basis. 2.2.2 Fuel consumption measurements Graduated conical glass flasks (Fisherbrand) were used for the determination of the fuel consumption of the machines (Figure 13). Three different sizes were used: 1000 ml, 500 ml and 250 ml. Figure 13: Filling flasks with diesel for measurements of fuel 27 2.2.3 Weight measurements A digital hanging scale (Voltcraft HS-50) was used to determine the weight composition of the product and byproduct of the maize ears with husk. The scale was fixed to a structure and the product was hooked on it as it is shown in Figure 14. Figure 14: Weighing set up during the assessment 2.2.4 Power consumption measurements The determination of the power consumption was performed by using digital watt meter (B+G E-Tech-DRT428BC). The device measures the actual load as well as the total power consumption and displays the data on an LCD screen. The values displayed on the power consumption measurement device show the active power of the machine. The displayed values switch every 5 seconds between the actual load and total consumption. The device was set in between the machine and the power supply by merging the connections in insulating screw joints. Before working in the electric circuit and installing the power consumption measurement device, a Voltcraft AC/DC Mini current clamp (VC-330) was used to verify that the electric circuit is interrupted (Figure 15). 28 Figure 15: Digital wattmeter for 3 phase powered machines (left) and the AC/DC Mini current clamp (right) 2.2.5 Global positioning system (GPS) data logger The GPS receiver (Renkforce GPS-102) was used to monitor the trajectory of machines during the land preparation and to measure the size of the fields. The leveling function of the GPS device was used to position the tractor horizontally to allow an accurate measurement of fuel in the tank (Figure 16). Figure 16: GPS tracker for leveling of the tractor 2.2.6 Sensor equipment summary Table 1 summarizes the equipment used during the experiments for each measurement parameter as well as displays accuracies and manufacturer. 29 Table 1: Summary of the equipment used during experiments Parameter Sensor model Accuracy (overall error) Company Power consumption DRT428BC 1% B+G ETech Current VC-330 2.5% Voltcraft GPS data GPS-102 Renkforce Weight HS-50 0.2 kg Voltcraft Weather data WS-1002-WIFI OBSERVER Temperature 2 F Humidity 5% Rainfall 10% Wind speed 10% Light 15% Air pressure +/- 3 hPa Conrad, Inc. 2.3 Methods 2.3.1 Weather data The installed weather station registers data in an interval of 5 minutes and saves them on an internal memory. These data can be copied manually on a micro SD card to be further processed. The weather data obtained are temperature, humidity, wind speed, dew point, wind chill, wind direction, absolute and relative pressure, rain rate, radiation, heat index and UV intensity. The data are analyzed and visualized with Microsoft Excel. 2.3.2 GPS interpretation During the field activities, the GPS sensor fixed to the tractor recorded the trajectories. Once the land preparation activities finished, the perimeter of the field was measured with the GPS sensor. GPS data were downloaded by using software CanWay (CanWay, version 1.1.12). Then the GPS information was exported as Keyhole Markup Language (KML) format to google maps4 under the usage of the My Maps function allowing the precise determination of the field area. 2.3.3 Questionnaire-based survey The energy input-output ratio of the 13 maize farms has been determined through observations and questionnaires (see Annex). The surveys were structured in three sections. The first part covers the basic information of the farmer, location of the farm, the farm size, cultivated crops as well as information about the available machinery. The second part comprises the production timeline to establish cropping calendars and the third part covers the energy inputs within farm activities. The questionnaires were conducted by two interviewers, mainly in French. 4 https://drive.google.com/open?id=1JB1amUoRbfCsayGYY9zDgDXHxWrZiK6z&usp=sharing 30 Interviewees were sampled with the snowball sampling method. The preselected, mechanized farmers helped to identify other maize farmers in the region, who would continue to identify others. A total of 6 interviews were held in Adjohoun and 7 interviews were held in Ktou. 2.3.4 Fuel consumption measurements Fuel consumption measurements were performed for non-stationary machines (tractors) and stationary machines (maize shellers and maize mills). The measurements at the non-stationary machines was carried out by comparing the fuel tank volume before and after the field activity. A wooden stick was used to measure the height of fuel in the tank. Figure 17 shows the process of refilling the tank of a tractor. Figure 17: Measuring fuel volume before filling the tank (left) and wooden stick used to control fuel volume (right) The fuel consumption was measured in Adjohoun for mulching, plowing and secondary tillage and in Ktou for plowing. Figure 18 shows the tractors used for land preparation. Table 2 shows the data of the tractors used at the farms in Adjohoun and Ktou. Figure 18: The tractors used in Adjohoun (left) and Ktou (right) during field preparation 31 Table 2: Data of the tractors used on the farms in Adjohoun and Ktou Farm Manufacturer Type Power (hp) Adjohoun Massey Ferguson 188 75 Ktou Mahindra 605 DI 60 For the measurements on the stationary machines, the operators emptied the fuel tank. The consumed fuel was calculated as the difference between fuel filled into the tank before the start of the machine and the amount of fuel which was left in the tank once the machine finished the process. The fuel consumption measurement of a maize sheller was performed in Ktou. A mobile shelling machine was powered by a small gasoline engine (G-Tech GX200), which was connected by a V-belt to the shelling unit. The gasoline engine has a nominated power of 6.5 hp. The gasoline motor has a hand-operated recoil starter. The maize ears were inserted in a funnel at the top of the machine. The shelling machine separates the maize grains from the cobs. The cobs were collected in a sack, the maize grains were collected in a large bowl as it is shown in Figure 19. Figure 19: Fuel measurement of the gasoline motor before and after operation (left) and shelling machine in operation (right) In Porto-Novo, two maize mills powered by a diesel engine were assessed. One of the maize mills was from the PTAA and the second mill was a public maize mill within the city of PortoNovo. In both cases, a diesel engine was connected by V-belt with the milling unit. Table 3 shows the specifications of the diesel engine at the PTAA. Table 3: Specifications of the diesel engine at PTAA Manufacturer Type Power R.P.M. S.F.C. IMEX V-Sensitive 5.9 kW 850 268 g / kWh 32 Maize is milled in batches. The milling machine was started by turning a starting crank to turn the flywheel to a certain speed. Once the machines started, maize grains were filled into the funnel. The operator sets the diameter between the rotating grinding disc and the fixed disc. The milled maize was collected in a bowl. This operation is carried out until the flour has a desired particle size. The milling process is shown in Figure 20. Figure 20: Milling process for maize grains at the PTAA. 2.3.5 Electricity consumption measurements In Porto-Novo, a maize mill powered by an electric engine was assessed. The electric engine is a Macforth three-phase induction motor. The specifications are shown in Table 4. Table 4: Specifications of the three-phase induction motor at PTAA Manufacturer Type Power R.P.M. Input Macforth Y 160M-4 11 kW 1460 440 V 19.5 A For the installation of the digital wattmeter, the machine was unplugged from the grid. Then the three phases were connected to the wattmeter and from there to the electric motor, as shown in Figure 21. The data of the electric consumption was registered before and after the run with the machine. The milling machine was operated in the same way as with the diesel engine. 33 Figure 21: Digital wattmeter connected to the electric motor 2.3.6 Energy Analysis The energy analysis of the maize value chain covers the total energy inputs in the maize production and their greenhouse gas emissions from land preparation until the end product. A comparison of farms with a high and low level of mechanization is performed. The data for the energy analysis is based from section 2.3.3, 2.3.4 and 2.3.5. Table 5 shows a summary of the data collected on the field trip and used for the energy and GHG emission analysis. Table 5: Overview of the data that has been collected and is used in the energy analysis Type Process Type of input Data from measurements Data from interviews Land preparation Clearing manual, machine machine hour, fuel man-hour Primary tillage: plow manual, machine machine hour, fuel man-hour Secondary tillage: disc harrow machine machine hour, fuel Secondary tillage: rotavator machine machine hour, fuel Plant production Planting manual man-hour, kg/ha Weeding manual man-hour Fertilizing manual man-hour, kg/ha Harvest manual man-hour, kg/ha Postharvest Peeling manual man-hour Shelling machine machine hour, fuel Milling machine machine hour, fuel 34 2.3.6.1 Energy and GHG emission conversion factors The energy and GHG emission analysis was done by using conversion factors, adapted from the approach of the Farm Energy Analysis Tool (FEAT) of the PennState University (Camargo et al., 2013). The mean conversion factors as assessed in literature are shown in Table 6. Table 6: Energy equivalents of agricultural inputs Type Unit Conversion factor (MJ/unit) Source Man-hour h 1.97 ( 0.02) 3, 6, 9, 10 Diesel fuel l 47.60 ( 7.14) 2, 3, 4, 5, 6, 9, 10 Gasoline fuel l 45.59 ( 2.46) 2, 10, 11, 12 Machinery h 192.73 ( 150.24) 6, 7, 14 Maize seeds kg 15.00 ( 0.42) 10, 13 N kg 65.09 ( 12.47) 1, 2, 3, 4, 5, 6, 7, 8 P2O5 kg 12.21 ( 3.75) 1, 2, 3, 4, 5, 6, 8 K2O kg 8.88 ( 2.82) 1, 2, 3, 4, 5, 6, 8 1 (West and Marland, 2002) 2 (Saiki et al., 1999) 3 (Houshyar et al., 2015) 4 (Manzone and Calvo, 2016) 5 (Reineke et al., 2013) 6 (Hamzei and Seyyedi, 2016) 7 (Kazemi et al., 2015) 8 (Pellizzi, 1992) 9 (Beheshti Tabar et al., 2010) 10 (Soni et al., 2013) 11 (Liu et al., 2013) 12 (Mandal et al., 2015) 13 (Jat et al., 2019) 14 (arauskis et al., 2018) The GHG emissions of the agricultural inputs are transformed into CO2 equivalents, which are shown in Table 7. Table 7: Greenhouse gas emissions in CO2 equivalents of agricultural inputs Type Unit CO2 equivalent (kg/unit) Source Diesel fuel l 2.92 ( 1.54) 1, 2, 3, 4, 5, 6 Gasoline fuel l 2.21 ( 2.01) 1, 3, 7 N kg 3.02 ( 2.25) 1, 2, 3, 4 P2O5 kg 0.71 ( 0.62) 1, 2, 3, 4 K2O kg 0.38 ( 0.28) 1, 2, 3, 4 1 (West and Marland, 2002) 2 (Cui et al., 2019) 3 (Lal, 2004) 4 (Wang et al., 2015) 5 (Houshyar et al., 2015) 6 (arauskis et al., 2018) 7 (Kazemi et al., 2015) 35 In the energy and GHG emission analysis farm inputs are converted with their corresponding conversion factors. Manual activities are converted using the conversion factor for man-hours. Also, the mechanized activities plowing, shelling and milling include the factor for man-hours to account for the time of the operators. The fuel consumption of the mechanized activities is converted with the conversion factors for diesel or gasoline. The fertilizers NPK and urea are converted using the percentages of the ingredients nitrogen, phosphorus and potassium and their corresponding conversion factors. The operation time of the tractor is multiplied by a conversion factor for machinery. The postharvest processes peeling, shelling and milling refer to the processing of the yield which is gained per hectare. 2.3.6.2 Energy use efficiency The energy use efficiency (EUE) is calculated by: ???????????? ?????? ???????????????????? = ???????????? ???????????? (???? h?? -1 ) ???????????? ?????????? (???? h?? -1) where the energy input includes all energy used for the maize production and the energy output considered the economic yield, which comprised the grain. The Specific energy (SE) is calculated by: ???????????????? ???????????? = ???????????? ?????????? (???? h?? -1 ) ?????????? ?????????? (???? h?? -1) where the energy input includes all energy used for the maize production up to the harvest and the maize yield (Yuan and Peng, 2017). 2.3.7 Designing and dimensioning of hybrid and PV system components In this study, the system components were sized for powering the milling machine. Three interventions were considered to assess the potential use of solar energy for a maize mill based on data gathered on the field trip in Benin. Table 8 shows the different system designs. The baseline is the currently most common mill that can be found in Benin, which is driven by a diesel motor. In intervention 1, the diesel motor is replaced with an electric motor, which is coupled to a diesel generator. In intervention 2, the mill is driven by an electric motor, which is powered by solar PV and batteries. In the third intervention, the mill is driven by an electric motor which is powered by a hybrid system consisting of diesel generator, solar PV and batteries. 36 Table 8: Overview of the system designs Name System type Main components Baseline Diesel system Diesel motor Intervention 1 Generator system Diesel generator, electric motor Intervention 2 Solar system Electric motor, batteries, PV panels Intervention 3 Hybrid system Diesel generator, electric motor, batteries, PV panels According to Wagner (2015), in the solar system, the dimensioning of the photovoltaic (PV) modules to cover a daily energy demand is calculated considering the incoming radiation of a given site using the following equation: ?????????? = ??0 ???? ?? ???? Where Ppeak is the peak power of the solar generator, E0 the irradiance at Standard Test Conditions (STC), Wd the daily energy requirement and G the global irradiation. The performance ratio (PR) describes the efficiency factor of the system, comprising the efficiency factor within charge controller, inverter and losses due to temperature, dust or degradation. In this study a PR of 0.75 was used. According to Wagner (2015), the capacity of the batteries required in a solar system can be calculated by: ???? = ???? ???? ???? ?????? Where CN is the capacity of the batteries in Ah, the days of autonomy is AF. Wd is the daily energy requirement in kWh, UN the nominal voltage of the batteries in V and DOD the depth of discharge of the batteries. However, the resulting capacity leads to an oversizing under real weather conditions, as the daily energy requirement is also partly covered by PV during days of low solar radiation. The batteries in this study were sized under the assumption that they could partly cover the daily energy demand to allow two days operation under days with low solar radiation. The days of lowest radiation in June, as measured with the weather station, are hence used for the battery sizing. The equation adapted in a way, that Wd is reduced by the daily energy generated by PV during these days. In this study a battery efficiency of 90% was used. 37 According to Ng et al. (2009), the state of charge (SOC) is defined as the releasable capacity Qreleasable of the battery relative to the rated capacity Qrated of the battery, which is given from the manufacturer: ?????? = ?????????????????????? ???????????? 100% The depth of discharge (DOD) is defined as the released capacity Qreleased of the battery relative to the rated capacity Qrated of the battery: ?????? = ?????????????????? ???????????? 100% The state of charge in this study is calculated with the coulomb counting method, where the time and current of charge and discharge is regarded. When the battery is charged and discharged over a period of time with a current Ib, the difference of the DOD in the operation period t is calculated by: ??????? = - ? ???? (??)???? ??0+t ??0 ???????????? 100% where Ib is positive for charging and negative for discharging. Charging and discharging comes by with efficiency losses, which are described with the factor . The depth of discharge after a certain time DOD(t) is the accumulation of DOD: ??????(??) = ??????(??0 ) + ?? ??????? the SOC of the battery after a certain period of time can hence be described as: ??????(??) = 100% - ??????(??) The fuel consumption of the diesel generator used in this study (Firman SDG7000SE) is at 0.35 l per kWh at rated power, according to the manufacturer. The design of the hybrid system was analyzed with weather data from Meteonorm. The hybrid system analyzed in this study uses as control strategy a load following dispatch. The para