Search

The estimation of lithium-ion batteries degradation plays an important role for the correct operation of smart grid and electric vehicle applications. In fact, only healthy battery energy storage systems meet minimum performance in terms of supplied voltage and power. Health prognostic is mandatory to ensure safe and reliable operation of batteries, as unsuccessful operation may cause technical/economic detriments or complete failures. The battery remaining useful life (RUL) depends on several random factors and thus it should be probabilistically characterized to allow the application of decision-making processes. In this article, a hybrid methodology is developed to characterize the RUL even with small amount of data, in view of the typical scarcity of data for adequately fitting the RUL probability distributions. Several distributions are considered in the methodology and compared for such purpose: the inverse Burr (IB) distribution, and mixtures of IB with inverse Gaussian and with inverse Weibull distributions. Expectation-maximization algorithms are specifically developed for the parameter estimation of the considered mixture distributions in the proposed methodology. The IB-based distributions are applied on a RUL dataset created by Monte Carlo sampling on an electrochemical battery model fitted upon given charge/discharge cycles. Numerical experiments are reported to assess the proposal with respect to benchmark distributions.

The availability of charging infrastructure is essential for large-scale adoption of electric vehicles (EV). Charging patterns and the utilization of infrastructure have consequences not only for the energy demand by loading local power grids, but influence the economic returns, parking policies and further adoption of EVs. We develop a data-driven approach that exploits predictors compiled from Geographic Information Systems data describing the urban context and urban activities near charging infrastructure to explore correlations with a comprehensive set of indicators that measure the performance of charging infrastructure. The best fit was identified for the size of the unique group of visitors (popularity) attracted by the charging infrastructure. Consecutively, charging infrastructure is ranked by popularity. The question of whether or not a given charging spot belongs to the top tier is posed as a binary classification problem and predictive performance of logistic regression regularized with an $\mathit {l}_{1}$ penalty, random forests and gradient boosted regression trees is evaluated. Obtained results indicate that the collected predictors contain information that can be used to predict the popularity of charging infrastructure. The significance of predictors and how they are linked with the popularity are explored as well. The proposed methodology can be used to inform charging infrastructure deployment strategies.

Cardiovascular diseases cause the death of almost 18 million people each year. Heart failure takes place when the heart does not pump a sufficient amount of blood to the body and is one of the most common causes of death. Healthcare technologies and Artificial Intelligence algorithms constitute promising solutions useful not only for the monitoring of people's health but also in order to tell which subjects are more prone to this problem, which is information of paramount relevance to save their lives. The goal of this paper is to understand the predictability of mortality of subjects suffering from left ventricular systolic dysfunction who previously experienced heart failures. To perform this important study, a publicly-available data set is considered that contains thirteen pieces of clinical, body, and lifestyle information about 299 subjects. In tackling this data set, not only do we wish to perform classification with reference to subjects' survival/death, but we also wish to automatically extract explainable knowledge about the reasons for the classification proposed. To this aim, we use DEREx, an Artificial Intelligence-based tool that relies on Evolutionary Algorithms and provides users with an easy-to-understand set of IF-THEN rules containing data set parameters. In this way, it performs the selection of the parameters that are the most relevant for the purpose of classification. We have run our experiments following a sound protocol established in the scientific literature for this data set. Our findings show that, apart from automatically obtaining easily interpretable knowledge, DEREx achieves better results in terms of widely-used quality indices as Matthews Correlation Coefficient, accuracy, and F score.

The operation of an Electric Arc Furnace (EAF) determines power quality disturbances in terms of voltage fluctuations and waveform distortions, due to the particular nature of the heating process and of the electric arc phenomenon. The analysis of DC EAFs and their induced power quality disturbances may be conveniently posed within a chaotic modeling framework. This paper presents a procedure for tuning the parameters of a chaotic model for the DC electrical arc. The procedure is based on a Simulink representation of the DC EAF installation and on the Chua's model to represent the arc chaotic behaviour. The parameters of the Chua's model are iteratively tested and optimized to minimize the errors in the evaluation of the spectral content of current waveforms at the point of common coupling. Data measured at the point of common coupling of an actual DC EAF are considered to carry out numerical simulations aimed at verifying the effectiveness of the proposed tuning procedure.

The aim of the paper is to develop a method for short-term deterministic and probabilistic photovoltaic (PV) power forecasting, and to present the results for an actual applicative case study. The proposed method is based on the exponential smoothing models, and two variants of this approach are analyzed within both the deterministic and the probabilistic frameworks. In particular, the probabilistic forecasting method is developed through an ensemble-based procedure that includes point forecasts obtained through several smoothing exponents. The case study discussed in this paper is based on experimental data obtained from an existing plant, and the corresponding results are compared with a benchmark based on the persistence model. It is found that the proposed exponential smoothing method significantly improves the result of the short-term PV power forecasting, both in the deterministic and probabilistic frameworks.

Lithium-ion (Li-ion) batteries are core equipment of Electric Vehicles (EVs) and they are expected to spread out through smart power systems (e.g., Smart Grids (SGs)). Batteries are subject to degradation over time, which gradually reduces the supplied voltage and capacity until they fail to meet minimum requirements. Battery prognostic is mandatory in order to keep track of the degradation over time and in order to schedule maintenance and replacement before failures, which can cause severe damages and harmful effects on EVs and SGs. However, the estimation of the State of Health (SoH) and of the Remaining Useful Life (RUL) of batteries is not trivial, since they cannot be directly measured. Predictive models should be therefore applied to accomplish this task, by linking SoH and/or RUL to variables measured during previous charge/discharge cycles (e.g., capacity, voltage, current, temperature). In this paper, a Multiple Linear Regression (MLR) model is developed to predict the SoH of batteries on the basis of historical data. The model exploits data collected for the battery under investigation and collected data of batteries characterized by Accelerated Degradation Tests (ADTs). Predictions of the RUL, given in terms of number of remaining charge/discharge cycles, can be further extracted from predictions of the SOH obtained through the proposal. Numerical experiments based on actual data are presented to validate the performance of the proposal.

This paper focuses on very-short-term photovoltaic power forecasting based solely on power measurement. The approach proposed in the paper is tailored for real time operation of smart grids and microgrids, that deals with energy management and control of the various resources of the grid and, particularly, on those based on renewable energy sources. These resources are characterized by variable power production, thus constituting a challenge for the effectiveness of strategies and tools of management and control. This paper refers to photovoltaic power production and presents a new real time forecasting method which improves the classical persistence model. The proposed method is based on the idea of conveniently weighting information on past data by imposing continuity of the function, of its first derivative and of the second derivative so obtaining three estimates of the function in the forecast interval which are then opportunely weighted to provide the forecast. The effectiveness of the proposed approach is shown in the numerical application based on actual measured data. The accuracy of the proposed approach is also tested through comparisons with two models based on the persistence and auto regressive moving average models.

Operating distribution networks by the dynamic thermal rating of lines and transformers allows for reaching the operational excellence and for optimizing the power delivery. However, due to the intrinsic uncertainties involved in the dynamic thermal rating estimation, the problem should be framed within a probabilistic environment. This paper focuses on distribution transformers; a novel non-parametric stress-strength model is presented in order to estimate the probability of the stress (i.e., the transformer loading current) to be smaller than the strength (i.e., the dynamic transformer rating). The model is based on a logistic regression of the companion binomial regression problem. Numerical experiments based on actual data collected at an Italian industrial facility are presented to estimate the performances of the stress-strength model.

This work presents a 2.4-GHz load-modulated balanced amplifier (LMBA) designed for a linear AM-PM characteristic. The balanced amplifier is designed in class-J to ensure high back-off efficiency, while the optimal load trajectory is found through experimental characterization of the amplifier in CW. When operated as an RF-input LMBA, the amplifier demonstrates 54 and 49% drain efficiency (DE) at peak output power and 6-dB back-off, respectively, with only 4 degrees of AM-PM variation across this power range.

This paper presents a dual-band power amplifier (PA) design approach that decouples the bias network and matching network designs despite widely-spaced operating bands. A broadband bias network is designed by applying filter techniques to present a high impedance at the fundamental through third harmonics of the two bands. Because this network does not load the device, the simplified real frequency technique can then be applied to design the dual-band matching network with reactive harmonic terminations. The demonstrator PA operates at 2.4 and 3.5 GHz with peak output power and drain efficiency of 41.5 dBm / 66% and 40.7 dBm / 58% respectively. The PA is characterized with a 10 MHz LTE signal, performing without digital predistortion with -38.5 dB ACLR at 33.1 dBm average output power at 2.4 GHz and -33.2 dB ACLR at 33.8 dBm average power at 3.5 GHz.

Kvalitu života diabetických pacientů lze zlepšit vytvořením personalizovaného kontrolního algoritmu integrovaného do umělé slinivky břišní, schopného dávkovat inzulín. Klíčovou akcí při stavbě tohoto umělého zařízení je vymyslet účinný algoritmus pro předpovídání budoucích hladin glukózy. V tomto článku je navržena evoluční strategie, tj. Algoritmus gramatické evoluce, k odvození personalizovaného předpovědního modelu pro hodnocení hodnot glykémie v budoucnosti na základ minulých měření glukózy a znalost bazálních a infuzních hladin inzulínu a spotřeby jídla. Cílem je objevit modely, které jsou nejen interpretovatelné ale také s nízkou složitostí pro použití v řídicím algoritmu, který je hlavním prvkem umělé slinivky břišní. K vyhodnocení navrhovaného evolučního automatického postupu byla použita databáze v reálném světě, kterou tvoří diabetičtí pacienti 1. typu. The quality of life of diabetic patients can be enhanced by devising a personalized control algorithm, integrated within an artificial pancreas, capable of dosing the insulin. A key action in the building of this artificial device is to conceive an efficient algorithm for forecasting future glucose levels. Within this paper, an evolutionary-based strategy, i.e., a Grammatical Evolution algorithm, is devised to deduce a personalized forecasting model to evaluate blood glucose values in the future on the basis of the past glucose measurements, and the knowledge of the basal and infused insulin levels and of the food consumption. The aim is to discover models that are not only interpretable but also with low complexity to be used within a control algorithm that is the main element of the artificial pancreas. A real-world database composed by Type 1 diabetic patients has been employed to evaluate the proposed evolutionary automatic procedure.

Blood pressure is a basic physiological parameter in the cardiovascular circulatory system. PhotoPlethysmoGraphy (PPG) signal represents a convenient, wearable, and low-cost technology that can be applied to various aspects of cardiovascular monitoring, including the detection of blood pressure, i.e., the hypertension level. The goal of this paper is to explore the behavior of a set of machine learning methods with respect to the hypertension risk stratification. With reference to this issue, the discrimination ability of the investigated algorithms has been considered at three different granularity levels, i.e. by suitably joining some of the classes making up the data set, so as to partition this latter into three different ways. To fulfill our goal, the Cuff-Less Blood Pressure Estimation Data Set is considered here. This data set is composed by many signals, including PPG data acquired from a group of subjects, and their Blood Pressure values, that are used to represent their hypertension levels. We have used a large group of machine learning tools, relying on differing working methods, and their numerical comparison has been carried out in terms of risk stratification results.

The management of diabetes is a very complex task, hence devising automatic procedures able to predict the glycemic level can represent a significant step towards the building of an artificial pancreas capable of providing the needed amounts of insulin boluses.This paper presents a Grammatical Evolution-based algorithm aiming at extrapolating a regression model able to estimate the blood glucose level in future instants of time through interstitial glucose measurements. The hypothesis is that the amounts of carbohydrates assumed, of basal insulin levels and of those administered with boluses are known. Experiments, performed on a real-world database made up of five patients suffering from Type 1 diabetes, are shown in terms of Clark Error Grid analysis. To evaluate the effectiveness of the predictions derived from the proposed approach, the results obtained are compared against those obtained by other state-of-the-art evolutionary-based methods very recently proposed.

This paper presents an overview of load modulation as an efficiency enhancement technique for radio frequency (RF) power amplifiers (PAs). A generalization of active load modulation is used to analyze different efficiency enhancement architectures such as the Doherty PA and the outphasing amplifier, providing qualitative insight of the techniques. Solutions to the challenges presented in load modulation PA design such as linearity, RF, baseband bandwidth and back-off efficiency enhancement are discussed through examples from the state-of-the art.

Forecasting the dynamic line rating allows to reach peaks of operational excellence upon electrical networks. Literature on dynamic rating has mainly addressed overhead lines, whereas lesser attention has been paid to buried cables. However, modeling the dynamics of the cable-soil system is quite a challenge, especially when both the day-ahead and intraday forecasting scenarios have to be considered in order to suit the usual operating tasks on electrical grids. This paper aims at providing a comparison among different forecasting methods, specially developed for such lead times. In particular, this paper: 1) develops a new physical-statistical method for intra-day forecasting scenarios; and 2) verifies the suitability of a data-driven method, which is an adaption of the state-of-the-art approach for dynamic overhead-line rating to the case of dynamic buried-cable rating, for both intraday and day-ahead scenarios. Numerical applications based on actual data are presented to validate the comparative study, and the forecasting results are compared with a naïve benchmark based on the persistence method.