Irrigation Systems, Rural Consumers, k-Means, Non-Technical Losses, Random Forest
Abstract
This work presents a proposal for the identification of non-technical losses in rural consumers directed to irrigated rice cultivation. The methodology employs in the structure of its algorithm, the k- Means and Random Forest methods, for the clustering and classification of data, using as inputs, monthly records of billed energy from a group of irrigating consumers in the county of Uruguaiana/RS. In addition, the algorithm proposed by this study performs the estimate of electricity consumption of rural consumers based on information on crop planting areas and variables of irrigation systems.
