Extended Isolation Forests for Fault Detection in Small Hydroelectric Plants

Rodrigo Barbosa De Santis,Marcelo Azevedo Costa

doi:10.3390/su12166421

Rodrigo Barbosa De Santis, Marcelo Azevedo Costa

Open Access

https://doi.org/10.3390/su12166421

Copy DOI

Journal: Sustainability	Publication Date: Aug 10, 2020
Citations: 34	License type: CC BY 4.0

Affiliation: Universidade Federal de Minas Gerais

Abstract

Maintenance in small hydroelectric plants is fundamental for guaranteeing the expansion of clean energy sources and supplying the energy estimated to be necessary for the coming years. Most fault diagnosis models for hydroelectric generating units, proposed so far, are based on the distance between the normal operating profile and newly observed values. The extended isolation forest model is a model, based on binary trees, that has been gaining prominence in anomaly detection applications. However, no study so far has reported the application of the algorithm in the context of hydroelectric power generation. We compared this model with the PCA and KICA-PCA models, using one-year operating data in a small hydroelectric plant with time-series anomaly detection metrics. The algorithm showed satisfactory results with less variance than the others; therefore, it is a suitable candidate for online fault detection applications in the sector.

Highlights

With energy demand expected to double by 2060, the development of clean energy sources is essential for guaranteeing an energy supply in the coming decades
The present paper proposes the application of isolation forest (iForest) and extended isolation Forest (EIF) to support fault detection and diagnosis of a hydro-generating unit (HGU) in an small hydroelectric plants (SHPs)
KICA-Principal Component Analysis (PCA) obtained the smallest difference between real fault detection, in general

Summary

Introduction

With energy demand expected to double by 2060, the development of clean energy sources is essential for guaranteeing an energy supply in the coming decades. Renewable energy already represents three quarters of yearly new installed capacity [1], and those related to water resources are the most applied. In this group, the construction of small hydroelectric plants (SHPs) has grown worldwide due to the lower initial investment, low operating costs, and increasing regulation of energy markets. Economic models of viability sensitivity analysis of SHPs stations are presented and applied to the energy context in Spain [9] and Greece [10]. A common factor among all these models of economic viability is the cost of operation and maintenance, which is a determining variable for the development of new stations

Methods

Results

Conclusion