We present here a combined aggregative short-term load forecasting method for smart grids, a novel methodology that allows us to obtain a global prognosis by summing up the forecasts on the compounding individual loads. More accurately, we detail here three new approaches, namely bottom-up aggregation (with and without bias correction), top-down aggregation (with and without bias correction), and regressive aggregation. Further, we have devised an experiment to compare their results, evaluating them with two datasets of real data and showing the feasibility of aggregative forecast combinations for smart grids.

To achieve the European climate targets and the Paris Agreements, at least 65% of the electricity needs to be generated from renewable energy sources by 2030. This requires a significant increase of distributed energy resources, posing a challenge for distribution system operators to integrate them into existing hierarchical grids. The concept of Local Flexibility Markets has recently gained attention as a market-based tool to tackle this challenge, making use of demand side flexibility. In this paper a Delphi method has been performed, showing that there are still numerous barriers in place preventing a widespread adoption of such markets in Europe. The main obstacles for market participants refer to standardisation issues. Based on that, a hybrid market model has been developed, comprising elements of a Local Flexibility Market and a Local Energy Market. To activate demand side flexibility from local energy transactions, spatio-temporally varying price signals are introduced, reflecting the constraints of the distribution grid. The paper shows, that this novel market approach helps to overcome relevant standardisation issues, but also certain barriers regarding end-users' lifestyles, which is because prices are comprehensible signals that can motivate end-users to participate. Moreover, a set of numerical examples is provided to illustrate the monetary benefits that could be gained by consumers and prosumers in the proposed hybrid market model. The examples show that the major share of the cost savings result from local energy trading, but the hybrid market model is also able to accumulate additional smaller revenues from providing flexibility. Finally, the systematic approach of characterising the market model in this paper offers a valuable framework for other researchers to map their ideas among existing approaches of Local Energy and Flexibility Markets.

The GreenSoul project introduces an innovative energy-efficient platform which enhances traditional public-use buildings with various technologies, such as smart adaptors, energy analyzers, an occupant aware decision support engine, mobile applications and interactive ambient interfaces. These enhancement aims to directly improve the interactions between occupants and energy-consuming assets in their environment. The GreenSoul framework is further enriched by the deployment of lightweight edge-computing GreenSoul-ed devices, which reduce energy consumption by cooperating with other devices, smart metering equipment and, very importantly, with eco-aware users. The decision making process is supported by a socio-economic behavioural model, which provides the necessary understanding of occupant indoor behaviour towards transforming energy consuming devices into active pro-sustainability agents that inform users how energy-efficiently they operate them, provide notifications as to improve that aspect, and even adapt their own functioning to reduce energy waste. These eco-wise devices, which are coined as GreenSoul-ed Things, are explained in detail throughout the presented work as well as the sensor-based architecture that supports their operation. The deployment of the framework across six pilot buildings is addressed, as well as the designed experimental setting to corroborate the potential of such a collaborative approach to enhance energy efficiency in office buildings.

The arrival of the smart grid paradigm has brought a number of novel initiatives that aim at increasing the level of energy efficiency of buildings such as smart metering or demand side management. Still, all of them demand an accurate load estimation. Short-term load forecasting in buildings presents additional requirements, among others the need of prediction models with simple or non-existing parametrisation processes. We extend a previous work that evaluated a number of algorithms to this end. Herewith we present several improvements including a variable data learning window and diverse learning data weighting combinations that further up improve our results. Finally, we have tested all the algorithms and modalities with four different datasets to show how the results hold up.