Event-Based Clustering with Energy Data

Abstract

This paper describes a stochastic clustering architecture that can be used for making predictions over energy data. The machine learning approach uses some new algorithms of hyper and frequency grids. The design is discrete, localised optimisations based on similarity, followed by a global aggregating layer. The global layer is entropy-based, allowing for a comparison with the recent distributed random neural network designs, for example. The topic relates to the IDEAS Smart Home Energy Project, where a client-side Artificial Intelligence component can predict energy consumption for appliances. The proposed data model is essentially a look-up table of the key energy bands that each appliance would use. Each band represents a level of consumption by the appliance, or the amount used in a time unit and the table can replace more complicated methods, usually constructed from probability theory, for example. Results show that the table can accurately disaggregate a single source to a set of appliances, because each appliance has quite a unique energy footprint. As part of predicting energy consumption, the model could possibly reduce costs by 50%, and more than that if proposed appliance schedules are also included. A second case study considers wind power patterns, where the grid optimises over the dataset columns in a self-similar way to the rows, allowing for some level of feature analysis.