SOM-based decision support system for reservoir operation management
The aim of this article is to analyze the different operation profiles for the Guadalmellato river reservoir and their relationships to rainfall and streamflow. The Viscovery SOMine model is used to revise the course of actions in specific situations in the past and provide decision support for future operation planning.
An ensemble empirical mode decomposition, self-organizing map, and linear genetic programming approach for forecasting river streamflow
The daily discharge at Dam Ten of the Kentucky River is forecasted employing linear genetic programming (LGP) in connection with ensemble empirical mode decomposition (EEMD) and self-organizing maps (SOM). The best results are obtained using a hybrid model, which uses EEMD for feature selection, Viscovery SOMs for clustering and LGP to calculate local prediction models on the clusters.
Self-organizing map (SOM) in wind speed forecasting: a new approach in computational intelligence (CI) forecasting methods
This article describes a wind-speed forecasting approach using the ordering capabilities of self-organizing maps. Viscovery SOMine is used to cluster 24-hour wind-speed profiles and predict wind speed on subsequent days for each cluster.
From smart meter data to pricing intelligence: real time BI for business innovation
Electricity consumption data is analyzed to uncover potentials for real-time BI, including demand response and pricing models. With Viscovery SOMine, clusters of similar consumers are found and analyzed.
Visual data mining: using self-organizing maps for electricity distribution regulation
Efficiency and cost data about Finnish electricity-distribution system operators from 2001–2004 is analyzed to provide efficiency benchmarks. Viscovery SOMine is used to cluster 356 records with information about annual costs, distributed energy, interruption times, number of users and network length.