Python-Based Planning Tool for LVAC Rural Electrification: Cluster Phase Balancing and Radial Topology

This paper focuses on open-source software-based optimal AC low voltage topology for rural electrification considering cluster phase balancing and radial topology (CB-RT) over a planning period of 30 years. Firstly, households are grouped into geography-based specific clusters using K-Means clustering and elbow methods. Secondly, mixed integer linear programming (MILP) is applied to balance phases in clusters in resulting an unbalanced current reduction. Then, radial topologies are proposed with different strategies which are: directed shortest path (CB-DSP), centered shortest path (CB-CSP), directed minimum spanning tree (CB-DMST), and centered minimum spanning tree (CB-CMST). Next, the sizing of protective devices (i.e., circuit breakers and fuses) is determined. Finally, the cluster phase balancing with these proposed radial topology strategies is compared based on the computation of the net present value (NPV) by taking into account the capital expenditure (CAPEX) and operation expenditure (OPEX) of the system. All simulation results are computed by Python-based open-source software. An AC low voltage distribution system in a rural village of Cambodia is chosen to validate the proposed methods.