Search

The rising awareness of environmental issues and the increase of renewable energy sources (RESs) has led to a shift in energy production toward RES, such as photovoltaic (PV) systems, and toward a distributed generation (DG) model of energy production that requires systems in which energy is generated, stored, and consumed locally. In this work, we present a methodology that integrates geographic information system (GIS)-based PV potential assessment procedures with models for the estimation of both energy generation and consumption profiles. In particular, we have created an innovative infrastructure that co-simulates PV integration on building rooftops together with an analysis of households’ electricity demand. Our model relies on high spatiotemporal resolution and considers both shadowing effects and real-sky conditions for solar radiation estimation. It integrates methodologies to estimate energy demand with a high temporal resolution, accounting for realistic populations with realistic consumption profiles. Such a solution enables concrete recommendations to be drawn in order to promote an understanding of urban energy systems and the integration of RES in the context of future smart cities. The proposed methodology is tested and validated within the municipality of Turin, Italy. For the whole municipality, we estimate both the electricity absorbed from the residential sector (simulating a realistic population) and the electrical energy that could be produced by installing PV systems on buildings’ rooftops (considering two different scenarios, with the former using only the rooftops of residential buildings and the latter using all available rooftops). The capabilities of the platform are explored through an in-depth analysis of the obtained results. Generated power and energy profiles are presented, emphasizing the flexibility of the resolution of the spatial and temporal results. Additional energy indicators are presented for the self-consumption of produced energy and the avoidance of CO2 emissions.

Abstract Nowadays, Non-Intrusive Load Monitoring techniques are sufficiently accurate to provide valuable insights to the end-users and improve their electricity behaviours. Indeed, previous works show that commonly used appliances (fridge, dishwasher, washing machine) can be easily disaggregated thanks to their abundance of electrical features. Nevertheless, there are still many ON/OFF devices (e.g. heaters, kettles, air conditioners, hair dryers) that present very poor power signatures, preventing their disaggregation with traditional algorithms. In this work, we propose a new online clustering method exploiting both operational features (peak power, duration) and external features (time of use, day of week, weekday/weekend) in order to recognize ON/OFF devices. The proposed algorithm is intended to support an existing disaggregation algorithm that is already able to classify at least 80% of the total energy consumption of the house. Thanks to our approach, we improved the performance of our existing disaggreation algorithm from 80% to 87% of the total energy consumption in the monitored houses. In particular, we found that 85% of the clusters were identified by only using operational features, while external features allowed us to identify the remaining 15% of the clusters. The algorithm needs to collect on average less than 40 operations to find a cluster, which demonstrates its applicability in the real world.

Spiking Neural Networks (SNNs), known for their potential to enable low energy consumption and computational cost, can bring significant advantages to the realm of embedded machine learning for edge applications. However, input coming from standard digital sensors must be encoded into spike trains before it can be elaborated with neuromorphic computing technologies. We present here a detailed comparison of available spike encoding techniques for the translation of time-varying signals into the event-based signal domain, tested on two different datasets both acquired through commercially available digital devices: the Free Spoken Digit dataset (FSD), consisting of 8-kHz audio files, and the WISDM dataset, composed of 20-Hz recordings of human activity through mobile and wearable inertial sensors. We propose a complete pipeline to benchmark these encoding techniques by performing time-dependent signal classification through a Spiking Convolutional Neural Network (sCNN), including a signal preprocessing step consisting of a bank of filters inspired by the human cochlea, feature extraction by production of a sonogram, transfer learning via an equivalent ANN, and model compression schemes aimed at resource optimization. The resulting performance comparison and analysis provides a powerful practical tool, empowering developers to select the most suitable coding method based on the type of data and the desired processing algorithms, and further expands the applicability of neuromorphic computational paradigms to embedded sensor systems widely employed in the IoT and industrial domains.

Due to the continuous increasing importance of renewable energy sources as an alternative to fossil fuels, to contrast air pollution and global warming, the prediction of Global Horizontal Irradiation (GHI), one of the main parameters determining solar energy production of photovoltaic systems, represents an attractive topic nowadays. Solar irradiance is determined by deterministic factors (i.e. the position of the sun) and stochastic factors (i.e. the presence of clouds). Since the stochastic element is difficult to model, this problem can benefit from machine learning techniques, like artificial neural networks. This work proposes a methodology to forecast GHI in short- (i.e. from 15 min to 60 min) and mid-term (i.e. from 60 to 120 min) time horizons. For this purpose, we designed, optimised and compared four neural network architectures for time-series forecasting, respectively based on: i) Non-Linear Autoregressive, ii) Feed-Forward, iii) Long Short-Term Memory and iv) Echo State Network. The original data-set, consisting of GHI values sampled every 15min, has been pre-processed by applying different filtering techniques. Our results analysis compares the performance of the proposed neural networks identifying the best in terms of error rate and forecast horizon. This analysis highlights that the clear-sky index results the preferred filtering technique by giving greatly improvements in data-set pre-processing, and Echo State Network gives best accuracy results.

In the Demand Side Management (DSM) context, residential customers have the potential for reducing costs and relieving the grid with non-thermostatic appliances. These appliances might be optimally scheduled by a central entity, taking into account user preferences. However, the user might not be able to communicate its preferences “a-priori”, leaving to the central entity the task of understanding preferences that should be learnt without causing discomfort to the user. With this premise, this study aims at exploring a DSM program that learns the acceptance of realistic simulated users to shift in time of home appliances, such as washing machines and dishwashers, analysing the benefits that arise from their inclusion. To this end, the proposed Acceptance Learning Algorithm 2.0 (ALA 2.0) minimises costs in scenarios with different energy sources and with a certain level of acceptance to shift in time, optimally scheduling the appliances according to the boundaries found by the proposed algorithm. ALA 2.0 is able to understand preferences also when modelling a behaviour of the user which is influenced by external factors not directly observable and when users make very few requests, interacting with the user in a simple way. Experimental results highlight that it is possible to understand the acceptance to the shift in time of the simulated users without any prior knowledge and without causing too much discomfort, achieving a win-win situation. As an example, more than 90% of requests were accepted in December, which is chosen as a representative month.

Hand gesture recognition applications based on surface electromiographic (sEMG) signals can benefit from on-device execution to achieve faster and more predictable response times and higher energy efficiency. However, deploying state-of-the-art deep learning (DL) models for this task on memory-constrained and battery-operated edge devices, such as wearables, requires a careful optimization process, both at design time, with an appropriate tuning of the DL models’ architectures, and at execution time, where the execution of large and computationally complex models should be avoided unless strictly needed. In this work, we pursue both optimization targets, proposing a novel gesture recognition system that improves upon the state-of-the-art models both in terms of accuracy and efficiency. At the level of DL model architecture, we apply for the first time tiny transformer models (which we call bioformers) to sEMG-based gesture recognition. Through an extensive architecture exploration, we show that our most accurate bioformer achieves a higher classification accuracy on the popular Non-Invasive Adaptive hand Prosthetics Database 6 (Ninapro DB6) dataset compared to the state-of-the-art convolutional neural network (CNN) TEMPONet (+3.1%). When deployed on the RISC-V-based low-power system-on-chip (SoC) GAP8, bioformers that outperform TEMPONet in accuracy consume 7.8×–44.5× less energy per inference. At runtime, we propose a three-level dynamic inference approach that combines a shallow classifier, i.e., a random forest (RF) implementing a simple “rest detector” with two bioformers of different accuracy and complexity, which are sequentially applied to each new input, stopping the classification early for “easy” data. With this mechanism, we obtain a flexible inference system, capable of working in many different operating points in terms of accuracy and average energy consumption. On GAP8, we obtain a further 1.03×–1.35× energy reduction compared to static bioformers at iso-accuracy.

Driving range is one of the most critical issues for electric vehicles (EVs): running out of battery charge while driving results in serious inconvenience even comparable to a vehicle breakdown, as an effect of long fuel recharging times and lack of charging facilities. This may discourage EVs for current and potential customers. As an effect, the dimensioning of the energy subsystem of an EV is a crucial issue: the choice of the energy storage components and the policies for their management should be validated at design time through simulations, so to estimate the vehicle driving range under reference driving profiles. Thus, it is necessary to build a simulation framework that considers an EV power consumption model that accounts for the characteristics of the vehicle and the driving route, plus accurate models for all power components, including batteries and renewable power sources. The goal of this paper is to achieve such an early EV simulation, through the definition of a SystemC-AMS framework, which models simultaneously the physical and mechanical evolution, together with energy flows and environmental characteristics. The proposed solution extends the state-of-the-art framework for the simulation of electrical energy systems with support for mechanical descriptions and the AC domain, by finding a good balance between accuracy and simulation speed and by formalizing the new information and energy flows. The experimental results demonstrate that the performance of the proposed approach in terms of accuracy and simulation speed w.r.t. the current state-of-the-art and its effectiveness at supporting EV design with an enhanced exploration of the alternatives.

Continuous glucose monitoring systems (CGMSs) allow measuring the blood glycaemic value of a diabetic patient at a high sampling rate, producing a considerable amount of data. These data can be effectively used by machine learning techniques to infer future values of the glycaemic concentration, allowing the early prevention of dangerous hyperglycaemic or hypoglycaemic states and better optimization of the diabetic treatment. Most of the approaches in the literature learn a prediction model from the past samples of the same patient, which needs extensive calibrations and limits the usability of the system. In this paper, we investigate the prediction models trained on glucose signals of a large and heterogeneous cohort of patients and then applied to infer future glucose-level values on a completely new patient. To achieve this purpose, we designed and compared two different types of solutions that were proved successful in many time-series prediction problems based respectively, on non-linear autoregressive (NAR) neural network and on long short-term memory (LSTM) networks. These solutions were experimentally compared with three literature approaches, respectively, based on feed-forward neural networks (FNNs), autoregressive (AR) models, and recurrent neural networks (RNN). While the NAR obtained good prediction accuracy only for short-term predictions (i.e., with prediction horizon within 30 min), the LSTM obtained extremely good performance both for short- and long-term glucose-level inference (60 min and more), overcoming all the other methods in terms of correlation between the measured and the predicted glucose signal and in terms of clinical outcome.

In recent years, designers are trying to move part of the computing tasks involved in Internet of Things applications from the cloud to the edge. This imposes increasing performance demands on edge nodes, which usually clash with their limited energy budget. An effective workaround is to leverage hardware capable of varying its computational precision at runtime, which can provide “good-enough” results while significantly reducing energy consumption. Dynamic Voltage and Accuracy Scaling (DVAS) and its variants are particularly promising methods to implement such hardware, due to their general applicability and contained overheads. However, these methods are negatively affected by the optimizations performed by commercial Electronic Design Automation (EDA) tools. As a consequence, when applied within a standard design flow, they do not yield the expected results. This paper describes a synthesis tool that solves this issue, allowing the integration of reconfigurable-precision circuits based on DVAS in standard design flows based on commercial tools. Moreover, our tool can receive information about the relevance of different precisions for the target application(s) that will use the circuit, and further optimize the design accordingly. When applied to two realistic use cases (neural network inference and image compression), our tool reduces the total energy consumption of reconfigurable-precision circuits of 20-25% compared to a straight-forward application of DVAS.

The evolution of industrial digitalisation has accelerated in recent years with the availability of hyperconnectivity, low-cost miniaturised electronic components, edge computing, and Internet of Things (IoT) technologies. More generally, with these key enablers, the concept of a system of systems (SoS) is becoming a reality in the industry domain. However, due to its complexity, the engineering process model adopted to design, develop, and manage IoT and SoS-based solutions for industry digitalisation is inadequate, inefficient, and frequently unable to manage the digitalisation solution’s entire life cycle. To address these limitations, we propose the Arrowhead Engineering Process (Arrowhead-EP) model and the Value Chain Engineering Process Map (VCEP-map), which explicitly reveal the interactions and dynamics of the engineering processes adopted by multistakeholder use cases in the industry domain. We decomposed and remodeled the engineering process to cover the complete life cycle of an industrial SoS, and we introduced a service-oriented solution intended to efficiently, flexibly, and effectively manage the three assets addressed by RAMI 4.0. The Arrowhead-EP model complemented by the VCEP-map fills the gaps identified in our literature-based analysis and satisfies the requirements of the life cycle management of a typical use case in the Industry 4.0 domain. In this regard, a specific example is used to illustrate the advantages of adopting the proposed engineering solution in a real multistakeholder use case.

SpiNNaker is a neuromorphic hardware platform, especially designed for the simulation of Spiking Neural Networks (SNNs). To this end, the platform features massively parallel computation and an efficient communication infrastructure based on the transmission of small packets. The effectiveness of SpiNNaker in the parallel execution of the PageRank (PR) algorithm has been tested by the realization of a custom SNN implementation. In this work, we propose a PageRank implementation fully realized with the MPI programming paradigm ported to the SpiNNaker platform. We compare the scalability of the proposed program with the equivalent SNN implementation, and we leverage the characteristics of the PageRank algorithm to benchmark our implementation of MPI on SpiNNaker when faced with massive communication requirements. Experimental results show that the algorithm exhibits favorable scaling for a mid-sized execution context, while highlighting that the performance of MPI-PageRank on SpiNNaker is bounded by memory size and speed limitations on the current version of the hardware.

The aging of rechargeable batteries, with its associated replacement costs, is one of the main issues limiting the diffusion of electric vehicles (EVs) as the future transportation infrastructure. An effective way to mitigate battery aging is to act on its charge cycles, more controllable than discharge ones, implementing so-called battery-aware charging protocols. Since one of the main factors affecting battery aging is its average state of charge (SOC), these protocols try to minimize the standby time, i.e., the time interval between the end of the actual charge and the moment when the EV is unplugged from the charging station. Doing so while still ensuring that the EV is fully charged when needed (in order to achieve a satisfying user experience) requires a “just-in-time” charging protocol, which completes exactly at the plug-out time. This type of protocol can only be achieved if an estimate of the expected plug-in duration is available. While many previous works have stressed the importance of having this estimate, they have either used straightforward forecasting methods, or assumed that the plug-in duration was directly indicated by the user, which could lead to sub-optimal results. In this paper, we evaluate the effectiveness of a more advanced forecasting based on machine learning (ML). With experiments on a public dataset containing data from domestic EV charge points, we show that a simple tree-based ML model, trained on each charge station based on its users’ behaviour, can reduce the forecasting error by up to 4× compared to the simple predictors used in previous works. This, in turn, leads to an improvement of up to 50% in a combined aging-quality of service metric.

One fundamental dimension in the design of an electrical energy system (EES) is the economic analysis of the possible design alternatives, in order to ensure not just the maximization of the energy output but also the return on the investment and the possible profits. Since the energy output and the economic figures of merit are intertwined, for an accurate analysis it is necessary to analyze these two aspects of the problem concurrently, in order to define effective energy management policies. This paper achieves that objective by tracking and measuring the energy efficiency and the cost effectiveness in a single modular framework. The two aspects are modeled separately, through the definition of dedicated simulation layers governed by dedicated virtual buses that elaborate and manage the information and energy flows. Both layers are simulated concurrently within the same simulation infrastructure based on SystemC-AMS, so as to recreate at runtime the mutual influence of the two aspects, while allowing the use of different discrete time scales for the two layers. Thanks to the tight coupling provided by the single simulation engine, our method enables a quick estimation of various cost metrics (net costs, annualized costs, and profits) of any configuration of EES under design, via an informed exploration of the alternatives. To prove the effectiveness of this approach, we apply the proposed strategy to two EES case studies, we explored various management strategies and the presence of different types and numbers of power sources and energy storage devices in the EES. The analysis proved to allow the identification of the optimal profitable solutions, thereby improving the standard design and simulation flow of EES.

Nearly 40% of primary energy consumption is related to the usage of energy in Buildings. Energy-related data such as indoor air temperature and power consumption of heating/cooling systems can be now collected due to the widespread diffusion of Internet-of-Things devices. Such energy data can be used (i) to train data-driven models than learn the thermal properties of buildings and (ii) to predict indoor temperature evolution. In this paper, we present a Grey-box model to estimate thermal dynamics in buildings based on Unscented Kalman Filter and thermal network representation. The proposed methodology has been applied in two different buildings with two different thermal network discretizations to test its accuracy in indoor air temperature prediction. Due to a lack of a real-world data sampled by Internet of Things (IoT) devices, a realistic data-set has been generated using the software Energy+, by referring to real industrial building models. Results on synthetic and realistic data show the accuracy of the proposed methodology in predicting indoor temperature trends up to the next 24 h with a maximum error lower than 2 °C, considering one year of data with different weather conditions.

The energy-optimal routing of Electric Vehicles (EVs) in the context of parcel delivery is more complicated than for conventional Internal Combustion Engine (ICE) vehicles, in which the total travel distance is the most critical metric. The total energy consumption of EV delivery strongly depends on the order of delivery because of transported parcel weight changing over time, which directly affects the battery efficiency. Therefore, it is not suitable to find an optimal routing solution with traditional routing algorithms such as the Traveling Salesman Problem (TSP), which use a static quantity (e.g., distance) as a metric. In this paper, we explore appropriate metrics considering the varying transported parcel total weight and achieve a solution for the least-energy delivery problem using EVs. We implement an electric truck simulator based on EV powertrain model and nonlinear battery model. We evaluate different metrics to assess their quality on small size instances for which the optimal solution can be computed exhaustively. A greedy algorithm using the empirically best metric (namely, distance × residual weight) provides significant reductions (up to 33%) with respect to a common-sense heaviest first package delivery route determined using a metric suggested by the battery properties. This algorithm also outperforms the state-of-the-art TSP heuristic algorithms, which consumes up to 12.46% more energy and 8.6 times more runtime. We also estimate how the proposed algorithms work well on real roads interconnecting cities located at different altitudes as a case study.

Finding the cost-optimal battery size in the context of parcel delivery with Electric Vehicles (EVs) requires solving a tradeoff between using the largest possible battery (so as to maximize the number of deliveries over a given time) and the relative costs (initial investment plus the unnecessary increase of the truck weight during delivery). In this paper, we propose a framework for the optimal battery sizing for parcel delivery with an electric truck; we implement an electric truck simulator including a nonlinear battery model to evaluate revenue, battery cost, charging cost, and overall profit for annual delivery. Our framework finds the cost-optimal battery size for different parcel weight distributions and customer location distributions. We analyze the effect of battery sizing on the profit, which is up to 56%.

Following the Smart City views, citizens, policy makers and energy distribution companies need a reliable and scalable infrastructure to manage and analyse energy consumption data in a city district context. In order to move forward this view, a city district model is needed, which takes into account dierent data-sources such as Building Information Models, Geographic Information Systems and real-time information coming from heterogeneous devices in the district. The Internet of Things paradigm is creating new business opportunities for low-cost, low-power and high-performance devices. Nevertheless, because of the "smart devices" heterogeneity, in order to provide uniform access to their functionalities, an abstract point of view is needed. Therefore, we propose an distributed software infrastructure, exploiting service-oriented middleware and ontology solutions to cope with the management, simulation and visualization of district energy data.

In order to systematically shift existing control and management paradigms in distribution systems to new interoperable communication supported schemes in smart grids, we need to map newly developed use cases to standard reference models like Smart Grid Architecture Model (SGAM). From the other side, any new use cases should be tested and validated ex-ante before being deployed in the real-world system. Considering various types of actors in smart grids, use cases are usually tested using co-simulation platforms. Currently, there is no efficient co-simulation platform which supports interoperability analysis based on SGAM. In this paper, we present our developed test platform which offers a support to design new use cases based on SGAM. We used this platform to develop a new scheme for wide area monitoring of existing distribution systems under growing penetration of Photovoltaic production. Off-the-shelf solutions of state estimation for wide area monitoring are either used for passive distribution grids or applied to the active networks with wide measurement of distributed generators. Our proposed distribution state estimation algorithm does not require wide area measurements and relies on the data provided by a PV simulator we developed. This practical scheme is tested experimentally on a realistic urban distribution grid. The monitoring results shows a very low error rate of about 1 % by using our PV simulator under high penetration of PV with about 30 % error of load forecast. Using our SGAM-based platform, we could propose and examine an Internet-of-Things-based infrastructure to deploy the use case.

In this paper we present VDJSeq-Solver, a methodology and tool to identify clonal lymphocyte populations from paired-end RNA Sequencing reads derived from the sequencing of mRNA neoplastic cells. The tool detects the main clone that characterises the tissue of interest by recognizing the most abundant V(D)J rearrangement among the existing ones in the sample under study. The exact sequence of the clone identified is capable of accounting for the modifications introduced by the enzymatic processes. The proposed tool overcomes limitations of currently available lymphocyte rearrangements recognition methods, working on a single sequence at a time, that are not applicable to high-throughput sequencing data. In this work, VDJSeq-Solver has been applied to correctly detect the main clone and identify its sequence on five Mantle Cell Lymphoma samples; then the tool has been tested on twelve Diffuse Large B-Cell Lymphoma samples. In order to comply with the privacy, ethics and intellectual property policies of the University Hospital and the University of Verona, data is available upon request to supporto.utenti@ateneo.univr.it after signing a mandatory Materials Transfer Agreement. VDJSeq-Solver JAVA/Perl/Bash software implementation is free and available at http://eda.polito.it/VDJSeq-Solver/.