Vegetation Management is a significant preventive maintenance expense in many power transmission and distribution companies. Traditional Vegetation Management operational practices have proven ineffective and are rapidly becoming obsolete due to the lack of frequent inspection of vegetation and environmental states. The rise of satellite imagery data and machine learning provides an opportunity to close the loop with continuous data-driven vegetation monitoring. This paper proposes an automated framework for monitoring vegetation along power lines using high-resolution satellite imagery and a semi-supervised machine learning algorithm. The proposed satellite-based vegetation monitoring framework aims to reduce the cost and time of power line monitoring by partially replacing ground patrols and helicopter or drone inspection with satellite data analytics. It is implemented and demonstrated for a power distribution system operator (DSO) in the west of Norway. For further assessment, the satellite-based algorithm outcomes are compared with LiDAR survey data collected by helicopters. The results show the potential of the solution for reducing the monitoring costs for electric utilities.

Risk assessment and management are some of the major tasks of urban power-grid management. The growing amount of data from, e.g., prediction systems, sensors, and satellites has enabled access to numerous datasets originating from a diversity of heterogeneous data sources. While these advancements are of great importance for more accurate and trustable risk analyses, there is no guidance on selecting the best information available for power-grid risk analysis. This paper addresses this gap on the basis of existing standards in risk assessment. The key contributions of this research are twofold. First, it proposes a method for reinforcing data-related risk analysis steps. The use of this method ensures that risk analysts will methodically identify and assess the available data for informing the risk analysis key parameters. Second, it develops a method (named the three-phases method) based on metrology for selecting the best datasets according to their informative potential. The method, thus, formalizes, in a traceable and reproducible manner, the process for choosing one dataset to inform a parameter in detriment of another, which can lead to more accurate risk analyses. The method is applied to a case study of vegetation-related risk analysis in power grids, a common challenge faced by power-grid operators. The application demonstrates that a dataset originating from an initially less valued data source may be preferred to a dataset originating from a higher-ranked data source, the content of which is outdated or of too low quality. The results confirm that the method enables a dynamic optimization of dataset selection upfront of any risk analysis, supporting the application of dynamic risk analyses in real-case scenarios.

Power utilities are continuously under high pressure to ensure the best performance of their grid. Nevertheless, power outages continue to be periodically observed. This paper assesses the applicability and implications of the Three-Phases method for optimized dataset selection in dynamic risk analysis, through a case study focusing on vegetation along power lines—a major hazard in power grid management. The case study comprises 17 different real-world datasets originating from 12 different types of data sources. We estimate how these datasets can inform eight parameters related to the physical configuration—one of the three dimensions impacting the probability of tree falls on power lines. The results provide two main take-aways: (1) datasets initially considered as less valuable for risk analysis can end up being the most relevant ones; (2) the potential of knowledge of a dataset needs to be assessed parameter per parameter. The results demonstrate that the Three-Phases method is a step toward traceable, data-driven, and dynamic risk analyses of power grids, resulting in a more reliable management of those large-scale infrastructures.