This is a Keynote Lecture delivered at the Big Data Analytics Summer School 2020 by Prof. Georgios Paliouras, NCSR “Demokritos”, iASiS Project Coordinator.

This module will cover the needs and challenges of distributed analytics and then dive into the details of scalable semantic analytics stack (SANSA) used to scalable analytics for knowledge graphs. This module will cover the setup, Apis and different layer of SANSA. At the end of this module, the audience will be able to execute examples and create programs that use SANSA APIs.

Transmission service operators (TSOs), regional security centres (RSCs), distribution service operators (DSOs), generation units (GUs) and balancing service providers (BSPs) need quick and reliable way of communication in order to secure power grid balance. They need to exchange information about grid stability, problems on the grid and defence plans in an easy and traceable way. In this paper we present software solution for handling these situations efficiently.

Numerous strategies were developed over the years in order to encourage users to reduce energy consumption and bolster energy efficiency. However, with increasing levels of efficiency achieved by most household appliances and high-consumption devices, one of the most impactful approaches that remains as a means to further increase energy efficiency is attempting to encourage users to behave in an energy-efficient manner. More precisely, positive behavior change can be motivated through the creation of unique social pressure and competition.

Intermittent renewable energy supply combined with electric and thermal energy storage technologies can cover the highest possible share of electricity, heating and cooling needs. However, their integration within the HVAC (Heating, Ventilation and Air-Conditioning) systems could result in far too complex installations, requiring intelligent energy management platforms for achieving their energy-efficient work.

Given the fact that renewable energy sources are increasing their share in the electricity market, in order to maintain the stable grid, i.e. match the production and the demand, it is crucial to have an accurate prediction of the expected accessible energy. Therefore, this paper is focused on providing the model for wind turbine production short-term forecast. The model is a deep neural network that includes LSTM, convolutional and dense layers, trained by the real-world data obtained from the wind farm in Krnovo, Montenegro.