Stefano Squartini

Optimal Home Energy Management Under Dynamic Electrical and Thermal Constraints

The optimization of energy consumption, with consequent costs reduction, is one of the main challenges in present and future smart grids. Of course, this has to occur keeping the living comfort for the end-user unchanged. In this work, an approach based on the mixed-integer linear programming paradigm, which is able to provide an optimal solution in terms of tasks power consumption and management of renewable resources, is developed. The proposed algorithm yields an optimal task scheduling under dynamic electrical constraints, while simultaneously ensuring the thermal comfort according to the user needs. On purpose, a suitable thermal model based on heat-pump usage has been considered in the framework. Some computer simulations using real data have been performed, and obtained results confirm the efficiency and robustness of the algorithm, also in terms of achievable cost savings.

Real-life voice activity detection with LSTM Recurrent Neural Networks and an application to Hollywood movies

A novel, data-driven approach to voice activity detection is presented. The approach is based on Long Short-Term Memory Recurrent Neural Networks trained on standard RASTA-PLP frontend features. To approximate real-life scenarios, large amounts of noisy speech instances are mixed by using both read and spontaneous speech from the TIMIT and Buckeye corpora, and adding real long term recordings of diverse noise types. The approach is evaluated on unseen synthetically mixed test data as well as a real-life test set consisting of four full-length Hollywood movies. A frame-wise Equal Error Rate (EER) of 33.2% is obtained for the four movies and an EER of 9.6% is obtained for the synthetic test data at a peak SNR of 0 dB, clearly outperforming three state-of-the-art reference algorithms under the same conditions.

A novel approach for automatic acoustic novelty detection using a denoising autoencoder with bidirectional LSTM neural networks

Acoustic novelty detection aims at identifying abnormal/novel acoustic signals which differ from the reference/normal data that the system was trained with. In this paper we present a novel unsupervised approach based on a denoising autoencoder. In our approach auditory spectral features are processed by a denoising autoencoder with bidirectional Long Short-Term Memory recurrent neural networks. We use the reconstruction error between the input and the output of the autoencoder as activation signal to detect novel events. The autoencoder is trained on a public database which contains recordings of typical in-home situations such as talking, watching television, playing and eating. The evaluation was performed on more than 260 different abnormal events. We compare results with state-of-the-art methods and we conclude that our novel approach significantly outperforms existing methods by achieving up to 93.4% F-Measure.

Online sequential extreme learning machine in nonstationary environments

Abstract System identification in nonstationary environments represents a challenging problem to solve and lots of efforts have been put by the scientific community in the last decades to provide adequate solutions on purpose. Most of them are targeted to work under the system linearity assumption, but also some have been proposed to deal with the nonlinear case study. In particular the authors have recently advanced a neural architecture, namely time-varying neural networks (TV-NN), which has shown remarkable identification properties in the presence of nonlinear and nonstationary conditions. TV-NN training is an issue due to the high number of free parameters and the extreme learning machine (ELM) approach has been successfully used on purpose. ELM is a fast learning algorithm that has recently caught much attention within the neural networks (NNs) research community. Many variants of ELM have been appeared in recent literature, specially for the stationary case study. The reference one for TV-NN training is named ELM-TV and is of batch-learning type. In this contribution an online sequential version of ELM-TV is developed, in response to the need of dealing with applications where sequential arrival or large number of training data occurs. This algorithm generalizes the corresponding counterpart working under stationary conditions. Its performances have been evaluated in some nonstationary and nonlinear system identification tasks and related results show that the advanced technique produces comparable generalization performances to ELM-TV, ensuring at the same time all benefits of an online sequential approach.

Multi-resolution linear prediction based features for audio onset detection with bidirectional LSTM neural networks

A plethora of different onset detection methods have been proposed in the recent years. However, few attempts have been made with respect to widely-applicable approaches in order to achieve superior performances over different types of music and with considerable temporal precision. In this paper, we present a multi-resolution approach based on discrete wavelet transform and linear prediction filtering that improves time resolution and performance of onset detection in different musical scenarios. In our approach, wavelet coefficients and forward prediction errors are combined with auditory spectral features and then processed by a bidirectional Long Short-Term Memory recurrent neural network, which acts as reduction function. The network is trained with a large database of onset data covering various genres and onset types. We compare results with state-of-the-art methods on a dataset that includes Bello, Glover and ISMIR 2004 Ballroom sets, and we conclude that our approach significantly outperforms existing methods in terms of F-Measure. For pitched non percussive music an absolute improvement of 7.5% is reported.

Unsupervised algorithms for non-intrusive load monitoring: An up-to-date overview

Research on Smart Grids has recently focused on the energy monitoring issue, with the objective to maximize the user consumption awareness in building contexts on one hand, and to provide a detailed description of customer habits to the utilities on the other. One of the hottest topic in this field is represented by Non-Intrusive Load Monitoring (NILM): it refers to those techniques aimed at decomposing the consumption aggregated data acquired at a single point of measurement into the diverse consumption profiles of appliances operating in the electrical system under study. The focus here is on unsupervised algorithms, which are the most interesting and of practical use in real case scenarios. Indeed, these methods rely on a sustainable amount of a-priori knowledge related to the applicative context of interest, thus minimizing the user intervention to operate, and are targeted to extract all information to operate directly from the measured aggregate data. This paper reports and describes the most promising unsupervised NILM methods recently proposed in the literature, by dividing them into two main categories: load classification and source separation approaches. An overview of the public available dataset used on purpose and a comparative analysis of the algorithms performance is provided, together with a discussion of challenges and future research directions.

Nonlinear speech enhancement: an overview

This paper deals with the problem of enhancing the quality of speech signals, which has received growing attention in the last few decades. Many different approaches have been proposed in the literature under various configurations and operating hypotheses. The aim of this paper is to give an overview of the main classes of noise reduction algorithms proposed to-date, focusing on the case of additive independent noise. In this context, we first distinguish between single and multi channel solutions, with the former generally shown to be based on statistical estimation of the involved signals whereas the latter usually employ adaptive procedures (as in the classical adaptive noise cancellation scheme). Within these two general classes, we distinguish between certain subfamilies of algorithms. Subsequently, the impact of nonlinearity on the speech enhancement problem is highlighted: the lack of perfect linearity in related processes and the non-Gaussian nature of the involved signals are shown to have motivated several researchers to propose a range of efficient nonlinear techniques for speech enhancement. Finally, the paper summarizes (in tabular form) for comparative purposes, the general features, list of operating assumptions, the relative advantages and drawbacks, and the various types of nonlinear techniques for each class of speech enhancement strategy.

An integrated system for voice command recognition and emergency detection based on audio signals

Emergencies and home automation commands are recognised by means of acoustic signals.Distress calls are recognised to allow tele-assistance by a relative or a care giver.A novelty detection algorithm detects abnormal acoustic events.The entire system has been implemented in a low-consuming embedded platform. The recent reports on population ageing in the most advanced countries are driving governments and the scientific community to focus on technologies for providing assistance to people in their own homes. Particular attention has been devoted to solutions based on acoustic signals since they provide a convenient way to monitor people activities and they enable hands-free human-machine interfaces. In this context, this paper presents a complete solution for voice command recognition and emergency detection based on audio signals entirely integrated in a low-consuming embedded platform. The system combines an active operation mode were distress calls are captured and a vocal interface is enabled for controlling the home automation subsystem, and a pro-active mode, were a novelty detection algorithm detects abnormal acoustic events to alert the user of a possible emergency. In the first operation mode, a Voice Activity Detector captures voice segments of the audio signal, and a speech recogniser detects commands and distress calls. In the pro-active mode, an acoustic novelty detector is employed in order to be able to deal with unknown sounds, thus not requiring an explicit modelling of emergency sounds. In addition, the system integrates a VoIP infrastructure so that emergencies can be communicated to relatives or care centres. The monitoring unit is equipped with multiple microphones and it is connected to the home local area network to communicate with the home automation subsystem. The algorithms have been implemented in a low-consuming embedded platform based on a ARM Cortex-A8 CPU. The effectiveness of the adopted algorithms has been tested on two different databases: ITAAL and A3Novelty. The obtained results show that the adopted solutions are suitable for speech and audio event monitoring in a realistic scenario.

Hybrid soft computing algorithmic framework for smart home energy management

Abstract Energy management in Smart Home environments is undoubtedly one of the pressing issues in the Smart Grid research field. The aim typically consists in developing a suitable engineering solution able to maximally exploit the availability of renewable resources. Due to the presence of diverse cooperating devices, a complex model, involving the characterization of nonlinear phenomena, is indeed required on purpose. In this paper an Hybrid Soft Computing algorithmic framework, where genetic, neural networks and deterministic optimization algorithms jointly operate, is proposed to perform an efficient scheduling of the electrical tasks and of the activity of energy resources, by adequately handling the inherent nonlinear aspects of the energy management model. In particular, in order to address the end-user comfort constraints, the home thermal characterization is needed: this is accomplished by a nonlinear model relating the energy demand with the required temperature profile. A genetic algorithm, based on such model, is then used to optimally allocate the energy request to match the user thermal constraints, and therefore to allow the mixed-integer deterministic optimization algorithm to determine the remaining energy management actions. From this perspective, the ability to schedule the tasks and allocate the overall energy resources over a finite time horizon is assessed by means of diverse computer simulations in realistic conditions, allowing the authors to positively conclude about the effectiveness of the proposed approach. The degree of realism of the simulated scenario is confirmed by the usage of solar energy production forecasted data, obtained by means of a neural-network based algorithm which completes the framework.

Optimal battery management with ADHDP in smart home environments

In this paper an optimal controller for battery management in smart home environments is presented in order to save costs and minimize energy waste. The considered scenario includes a load profile that must always be satisfied, a battery-system that is able to storage electrical energy, a photovoltaic (PV) panel, and the main grid that is used when it is necessary to satisfy the load requirements or charge the battery. The optimal controller design is based on a class of adaptive critic designs (ACDs) called action dependent heuristic dynamic programming (ADHDP). Results obtained with this scheme outperform the ones obtained by using the particle swarm optimization (PSO) method.