Keywords
Pre-processing, clustering, post-processing, spatio-temporal patterns, data science, energy
Start Date
27-6-2018 9:00 AM
End Date
27-6-2018 10:20 AM
Abstract
In this paper, we aim at assessing the complexity and resilience of the European Transmission Power Grid (ETPG) following a data science approach. We consider open data related to energy policies and infrastructural and economic variables, together with ETPG reliability data (i.e., major failures and blackout data) of most European countries, considering data from a period of 14 years’ (2002 – 2014). A Data Science approach is used to understand spatio-temporal patterns of failures and blackouts of the ETPG along the different countries and periods. A combination of clustering methods with post-processing interpretation techniques and complex networks analysis is applied to understand the factors associated with blackouts and failures in the different regions and temporal periods. An innovative approach in the field of multivariate time series is used to introduce additional covariables into the analysis, by completing the blackout data with additional open data related to energy policies and infrastructural and economic variables. Adding contextual information to time series contribute to a better understanding of the phenomenon. Our results offer a novel approach to understand the relation between these variables and to improve our ability to maintain and guarantee the ETPG’s resilience, defined by its structural integrity, security of supply and transport efficiency.
A data science approach to assess resilience and complexity in the European Transmission Power Grid
In this paper, we aim at assessing the complexity and resilience of the European Transmission Power Grid (ETPG) following a data science approach. We consider open data related to energy policies and infrastructural and economic variables, together with ETPG reliability data (i.e., major failures and blackout data) of most European countries, considering data from a period of 14 years’ (2002 – 2014). A Data Science approach is used to understand spatio-temporal patterns of failures and blackouts of the ETPG along the different countries and periods. A combination of clustering methods with post-processing interpretation techniques and complex networks analysis is applied to understand the factors associated with blackouts and failures in the different regions and temporal periods. An innovative approach in the field of multivariate time series is used to introduce additional covariables into the analysis, by completing the blackout data with additional open data related to energy policies and infrastructural and economic variables. Adding contextual information to time series contribute to a better understanding of the phenomenon. Our results offer a novel approach to understand the relation between these variables and to improve our ability to maintain and guarantee the ETPG’s resilience, defined by its structural integrity, security of supply and transport efficiency.
Stream and Session
Stream B: (Big) Data Solutions for Planning, Management, and Operation and Environmental Systems
B3: Sixth Session on Data Mining as a Tool for Environmental Scientists (S-DMTES-2018)