Networked rural electrification – optimal network design under complex topography

Li, Jerry Chun-Fung, author
Young, Peter, advisor
Zimmerle, Daniel, advisor
Cale, Jim, committee member
Cross, Jeni Eileen, committee member
Journal Title
Journal ISSN
Volume Title
The 7th of United Nations' Sustainable Development Goals (SDG7) aims to "ensure access to affordable, reliable, sustainable and modern energy for all" by 2030. While substantial progresses have been made in the last few years, 759 million people in rural areas still have no or limited access to electricity. Due to the distances and geographical complexity of rural areas, providing electricity to this unserved population is very costly. As IEA recently pointed out, rural electrification is increasingly costly. With the current electrification approach, it is expected that 660 million people will remain without electricity access by 2030. In addition, accurate planning for small rural power system is difficult as both demand and energy resource forecasts are highly uncertain. Thus, achieving SDG7 is very challenging. In this research, a Networked Rural Electrification framework has been proposed. This approach can potentially accelerate SDG7 by reducing system cost, enhancing reliability, and offering installation flexibility for small communities in remote areas. In this framework, villages and generation facilities are connected via an optimal, low voltage network that can be built with inexpensive poles and cables. To make this approach economically feasible, cost for building the network is crucial. A specific difficulty associated with this approach is the anisotropicity of search space for optimal design of the power distribution network, which results from complex topographical variations in these rural areas. Traditional optimization methods are not suitable for designing this network because of computational complexity, accuracy requirement, and practical implementation considerations. To address the issues, new computation methods and tools have been developed. These include (i) Multiplier-accelerated A* (MAA*) and (ii) Adaptive Multiplier-accelerated A* (AMAA*) algorithms, which resolve the computational complexity problem by significantly reduce computation time while maintaining good optimality, and (iii) Levelized Interpolative Genetic Algorithm (LIGA) which, when used in conjunction with A*, MAA*, or AMAA*, provides viable alternative plans to tackle unexpected route change problem right before or even during project implementation, and (iv) a fuzzy rule-based system for further network topology optimization.
2022 Spring.
Includes bibliographical references.
Rights Access
genetic algorithm
optimal path-finding
optimal network
fuzzy inference
rural electrification
Associated Publications