Development and validation of a systems-level cost optimization tool for solar-powered drip irrigation systems for smallholder farms
Author(s)Grant, Fiona(Fiona R.)
Massachusetts Institute of Technology. Department of Mechanical Engineering.
Amos G. Winter, V.
MetadataShow full item record
Drip irrigation is a micro-irrigation technology that has been shown to conserve water and significantly increase crop yield. This technology could be particularly beneficial to the world's estimated 500 million smallholder farmers, but these systems tend to be financially inaccessible to this population. Drip systems require costly components including a pipe network, emitters, a pump and a power system; due to limited access to electricity, many smallholder farmers would require off-grid solutions. Designing reliable, low cost, off-grid drip irrigation systems for smallholder farms could significantly reduce the barrier to adoption. This thesis presents a comprehensive model that holistically simulates the system behavior and cost-optimizes the system design. A custom, low pressure emitter is used in the hydraulic network of these systems. This design tool produces low cost, solar powered drip irrigation systems that are tuned for a specific geographic location and crop type.The model simulates the agronomic, hydraulic, pump and power system behaviors. A PSO algorithm is used along with local economic data to optimize for either the lowest cost design or the design that produces the highest profit from crop yield over the lifetime of the system. The design must include a properly sized pump and solar panels, and may include energy storage in the form of a battery, a tank or both. The reliability of the design is assessed by simulating its performance over a growing season using local weather and crop data. The extent of the model is explored through sensitivity analysis and a series of sample cases for field sizes ranging from 0.125 to 2 ha. It is shown that the optimization can reduce the life cycle cost of a system by 62% compared to the conventional method for sizing solar powered drip irrigation systems. The potential of the model to inform emitter design and pump selection is also explored.The simulation portion of the model is validated through a set of field trials where two solar powered drip systems were installed and run on small farms in Jordan and Morocco for a full growing season. Future iterations of the model will include an optimized hydraulic network layout and irrigation operation scheme, as well as more flexible pump selection criteria. Future field work will validate the optimized operation scheme, which will be used along with feedback from the farmers to design a custom controller.
Thesis: S.M., Massachusetts Institute of Technology, Department of Mechanical Engineering, 2019Cataloged from PDF version of thesis.Includes bibliographical references (pages 109-113).
DepartmentMassachusetts Institute of Technology. Department of Mechanical Engineering
Massachusetts Institute of Technology