Apostolos Vavouris

Emotion Artificial Intelligence (AI) is transforming the capabilities of personal assistant by enabling real-time adaptation to user emotions, behaviours, and contextual needs. This paper explores the potential of emotion-adaptive eco-feedback in personal assistant, particularly within home environments, to foster well-being, energy efficiency, and personalised user experiences. Currently, there is limited research on how users perceive emotion-adaptive eco-feedback and how emotion AI can be adopted in the eco-feedback within personal assistant in real-world settings. To address this, we employed a co-design method — Matchmaking for AI — to facilitate collaboration between real users and researchers. We built a living lab with 11 participants in Germany for half a year and conducted two experimental sessions: a pre-interview to understand user behaviours, requirements, and expectations on eco-feedback, and a co-design session using matchmaking for AI after half a year based on their appliance energy consumption data collected by our smart plugs using our open. DASH platform. The co-design sessions collaboratively brainstorm ideas for potential emotion AI adoptions and identify what their needs should be addressed by emotion AI technology. Through a co-design session, we generated eight design ideas that integrate emotion AI into eco-feedback. These concepts include emotion-adaptive eco-feedback framing, emotion-timed interaction and delivery and emotion-aware environment and social adaption. Our work explores the potential of using Emotion AI in eco-feedback within personal assistant and also provides new insights into AI co-design methodologies.

The industrial sector, which is characterised by complex and diverse equipment and processes, is a core contributor to carbon emissions. Agriculture, in particular, has seen an increase in carbon dioxide emissions, in contrast to methane and nitrous oxide emissions, due to the increased use of novel, energy-intensive agricultural technology. Therefore, to meet Net Zero targets, analyses of the energy consumption of dairy technologies are necessary to enhance the energy efficiency of the agricultural equipment. Incidentally, the only public electrical measurements datasets of agricultural technology are one for poultry feed for 111 days, one for a synthetic hourly load profile of a dairy farm in Finland and another with aggregate readings and submetered milking robots from three farms in Germany. This dataset aims to accelerate energy efficiency and management services in the sector by curating and providing access to a comprehensive, first-of-its-kind, granular (1-sec) power readings dataset from 30 dairy farms across Germany with 72 submetered energy-intensive equipment, for a period of over one year.

Emerging concepts for sustainable urban development, such as Positive Energy Districts/Neighbourhoods (PEDs/PENs), are being implemented in Europe in light of the goal of net-zero carbon emissions by 2050. From the building to the district/neighbourhood scales, these projects are fostering the implementation of local Renewable Energy Sources (RES), encouraging the adoption of Electric Vehicles (EVs) and Smart Home Technologies (SHTs). In this context, a systemic analysis of energy prosumption in these new urban areas is of paramount importance to evaluate their social and technical impacts. This dataset from a Norwegian PEN enables such systemic analysis by bringing the following quantitative and qualitative data: (i) the aggregate loads of six households, (ii) the estimated energy production based on the installed solar panel capacity, (iii) the labelling of individual activities, (iv) solar irradiance and weather data, (v) time of use of electric appliance surveys, and (vi) in-depth semi-structured interviews on household energy practices. Therefore, this dataset facilitates, multi-disciplinary research on the sociotechnical aspects of PENs and net-positive buildings for their design and implementation, helping to answer the “what”, “why”, and “how” of energy prosumption.

Charging electric vehicles (EVs) with renewable energy can lessen their environmental impact. However, the fluctuating availability of renewable energy affects the sustainability of public EV charging stations. Nearby public charging stations may utilize differing energy sources due to their microgrid connections - ranging from exclusively renewable to non-renewable or a combination of both - highlighting the substantial variability in energy supply types within short distances. This study investigates the near-future scenario of integrating dynamic renewable energy availability in charging station navigation to impact the choices of EV users towards renewable sources. We conducted a within-subjects design survey with 50 car users and semi-structured interviews with 10 EV users from rural, suburban, and urban areas. The results show that when choosing EV charging stations, drivers often prioritize either time savings or money savings based on the driving scenarios that influence drivers’ consumer value. Notably, EV users tend to select renewable-powered stations when they align with their main priority, be it saving money or time. This study offers end-user insights into the front-end graphic user interface and the development of the back-end ranking algorithm for navigation recommender systems that integrate dynamic renewable energy availability for the sustainable use of electric vehicles.

Electrification of transportation is gaining traction as a viable alternative to vehicles that use fossil-fuelled internal combustion engines, which are responsible for a major part of carbon dioxide emissions. This global turn towards electrification of transportation is leading to an exponential energy and power demand, especially during late-afternoon and early-evening hours, that can lead to great challenges that electricity grids need to face. Therefore, accurate estimation of Electric Vehicle (EV) charging loads and time of use is of utmost importance for different participants in the electricity markets. In this paper, a scalable methodology for detecting, from smart meter data, household EV charging events and their load consumption with robust evaluation, is proposed. This is achieved via a classifier based on Random Decision Forests (RF) with load reconstruction via novel post-processing and a regression approach based on sequence-to-subsequence Deep Neural Network (DNN) with conditional Generative Adversarial Network (GAN). Emphasis is placed on the generalisability of the approaches over similar houses and cross-domain transferability to different geographical regions and different EV charging profiles, as this is a requirement of any real-case scenario. Lastly, the effectiveness of different performance and generalisation loss metrics is discussed. Both the RF classifier with load reconstruction and the DNN, based on the sequence-to-subsequence model, can accurately estimate the energy consumption of EV charging events in unseen houses at scale solely from household aggregate smart meter measurements at 1–15 min resolutions.

In body-centric communications, energy efficiency is a critical performance metric, while the achievable data rate is not of primary concern. In this paper we present a novel modulation scheme, which can be efficiently used in body-centric terahertz (THz) nanonetworks. The proposed scheme is a combination of the time-spread On-Off keying (TS–OOK) and the pulse position modulation (PPM) and presents lower energy consumption, compared to other existing methods as TS–OOK, at a minor cost to the data rate. Furthermore, another important aspect is that the proposed modulation scheme can be effectively used to mitigate the impact of the specific kind of noise in THz body-centric communications, thus leading to better error performance. Finally, we present analytical and simulation results in order to compare the new scheme with the existing TS–OOK.