Given the ubiquity of mobile devices, understanding the context of human activity with non-intrusive solutions is of great value. A novel deep neural network model is proposed, which combines feature extraction and convolutional layers, able to recognize human physical activity in real-time from tri-axial accelerometer data when run on a mobile device. It uses a two-layer convolutional neural network to extract local features, which are combined with 40 statistical features and are fed to a fully-connected layer. It improves the classification performance, while it takes up 5–8 times less storage space and outputs more than double the throughput of the current state-of-the-art user-independent implementation on the Wireless Sensor Data Mining (WISDM) dataset. It achieves 94.18% classification accuracy on a 10-fold user-independent cross-validation of the WISDM dataset. The model is further tested on the Actitracker dataset, achieving 79.12% accuracy, while the size and throughput of the model are evaluated on a mobile device.

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.

Alternating current (AC) microgrids are expected to operate as active components within smart distribution grids in the near future. The high penetration of intermittent renewable energy sources and the rapid electrification of the thermal and transportation sectors pose serious challenges that must be addressed by modern distribution system operators. Hence, new solutions should be developed to overcome these issues. Microgrids can be considered as a great candidate for the provision of ancillary services since they are more flexible to coordinate their distributed generation sources and their loads. This paper proposes a method for compensating microgrid power factor and loads asymmetries by utilizing advanced functionalities enabled by grid tied inverters of photovoltaics and energy storage systems. Further, a central controller has been developed for adaptively regulating the provision of both reactive power and phase balancing services according to the measured loading conditions at the microgrid’s point of common coupling. An experimental validation with a laboratory scale inverter and a real time hardware in the loop investigation demonstrates that the provision of such ancillary services by the microgrid can significantly improve the operation of distribution grids in terms of power quality, energy losses and utilization of available capacity.

Recent decentralization of electricity systems together with the decarbonization and several changing societal demands are giving rise to different application scenarios such as microgrids. A microgrid is a small-scale electrical system which consists of several loads and sources (conventional and renewables) that can either operate autonomously in a stand-alone mode or interconnected with the main grid. The design and development of such a smart microgrid in a university campus is proposed within the 3DMicroGrid project (funded through the ERANETMED European Union's initiative). This paper reviews the main components and characteristics of similar microgrids developed around the world. Furthermore, this study provides the design guidelines, the main functionalities, the key components and the control architecture for developing the microgrid proposed by the 3DMicroGrid project. A simulation model has been developed and initial results are demonstrated for the operation of this microgrid. The recommendations and insights are replicable to any solar priority country for future microgrids pilots.