Antonio Bracale

Forecasts of waveform harmonic distortions can be used by operators to implement corrective actions for complying with suggested limits, to refine the components’ dynamic rating predictions, and to run scheduling strategies for controllable resources in smart/micro grids. Nevertheless, developing a unified forecasting methodology for all the harmonic components is challenged by their different time-varying characteristics. To tackle this problem, this paper proposes to integrate clustering techniques in probabilistic current and voltage harmonic forecasting methodologies. Identifying common characteristics in the harmonic component patterns enables building the underlying forecasting models in a diversified manner, with a model for each cluster. Since true memberships of unobserved patterns to clusters are unknown to the forecaster, the individual forecasts must be optimally recombined to get the final predictions. This paper proposes the combination through an optimized Beta-transformed Linear Pool (BLP), mitigating the lack of calibration that can occur when combining probabilistic forecasts. To assess the clustering effectiveness, parametric and nonparametric underlying forecasting models are considered, and different combination benchmarks are taken as reference. Applications based on actual measurements in commercial, office, or residential sites confirm the suitability of the proposal, with relative improvements of the forecast accuracy over persistence benchmarks from 2 to 48%.

In this paper, the aggregated modelling of large-scalewind farms for harmonic studies is considered. The limitations of the harmonic emission assessment based on the IEC 61000-3-6 summation rule are analysed in a probabilistic framework. An iterative probabilistic procedure based on the Quasi Monte Carlo technique is proposed and applied to a simplified electromagnetic transient model of an exemplary wind power plant. Then, detailed results obtained are used to quantify the above mentioned limitations of the summation rule.

Secondary and primary substations of networks with electric vehicle (EV) chargers and photovoltaics (PVs) experience net loads characterized by uncertainty. Accurate characterization of EV, PV and net load energy profiles is necessary to plan new installations and to develop forecasting methodologies. This paper provides a novel contribution to the energy profile characterization of EVs and PVs, exploiting clustering techniques in a hierarchical framework to eventually characterize the overall net load profiles. In the proposal, the lower levels of the hierarchy identify clusters of EV load and PV generation profiles at individual installations, alternatively using one clustering technique among DBSCAN, Gaussian mixture models (GMMs), K‐means algorithm (KMA), and spectral clustering (SC). The intermediate levels of the hierarchy reconstruct the overall EV load and PV generation profiles through a proposed frequentist combination of the lower‐level profiles. The upper level of the hierarchy characterizes the overall net load through a novel approach based on the quantile convolution of the intermediate‐level EV and PV profiles. Real EV load and PV generation data are used to evaluate the performance of the presented hierarchical methodology, with relative fitting improvements between 1% and 8% (compared to a two‐level hierarchical benchmark) and between 16% and 29% (compared to a direct, non‐hierarchical benchmark).

The analysis of power quality disturbances in distribution systems has gained significance with the diffusion of electric vehicles (EVs). Waveform distortions are interesting since EV currents introduce distortions with spectral components in both low and high-frequency bands. This paper develops specific indices to assess cumulative emissions from single-phase EV on-board chargers, extending the aggregation and diversity factors to the supra-harmonic range. The methodology accounts for variables such as EV charging powers, upstream network impedance, and number of EVs. A simplified time-domain model of a low-power unidirectional converter, commonly used for EV battery charging, is employed to balance circuit complexity and computational effort. This model allows for sensitivity analyses of key parameters influencing charger emissions. Numerical applications are carried out for both individuals and groups of EV chargers at a charging station. Results highlight the need for careful quantification of aggregated EV emissions, showing that supra-harmonic emissions are highly sensitive to variations in the power absorbed by EV chargers. Notably, their cumulative impact is much lower when chargers operate at different power levels than when all chargers operate at the same power level. These findings underscore the importance of accurately assessing the impact of EV charging on power quality.

The proliferation of photovoltaic (PV) systems connected to low voltage (LV) distribution networks can have detrimental impacts on waveform distortion. This is caused by the power electronic interface, with voltage source converter (VSC) technology being by far the most prevalent. As such, proper emission models, which can account for the non‐linear operation of VSCs used in PV systems have to be developed and investigated for the assessment of harmonic distortion in LV networks. This paper compares different frequency domain models (FDMs), specifically, methods based on frequency coupling matrices and an analytical method based on a harmonically coupled impedance matrix, for the type of single‐phase PV systems typically found in LV distribution networks. Two case studies are presented to compare the models in terms of computational complexity and accuracy, with results showing that models accounting for the interaction between same order harmonics are sufficiently accurate.

Inverter-based resources (IBRs) are a new type of generators entering power systems. Industries are concerned about the harmonic impact caused by IBR plants. How to model IBRs for harmonic studies has, therefore, become an important topic. This industry application-oriented tutorial paper presents a comprehensive review and analysis of the harmonic behaviors of IBR units, covering their harmonic characteristics, advanced and practical harmonic models, methods for model parameter determination and more. Lab test results on a physical IBR unit are presented to substantiate the findings. It is hoped this paper will clarify the confusion between the harmonic models of VSC (voltage source converter) based IBR units versus LCC (line commutated converter) based loads. Furthermore, a comparison between IBRs and synchronous generators reveals many similarities in their harmonic behaviors.

Waveform distortion is one of the most common disturbances in power quality in real power grids. Accurate assessment of the distortion level for voltage and current is a challenge for researchers, as a result of simultaneous occurrence of a large number of time-varying spectral components and the problem of proper synchronization of measurement window. The article presents a new method for assessing low-frequency harmonics (LFH) and high-frequency harmonics (HFH), which are supraharmonics with a frequency that is a multiple of fundamental frequency. The proposed method is based on discrete Fourier transform analysis using a “flat-top” window, where the sliding measurement window is synchronized with the short-term fundamental period by using empirical Fourier decomposition. The presented approach focuses on increasing the accuracy of assessment of LFH and HFH in power grids only by modifying the synchronization of measurement window. The presented approach makes significant modifications to the current IEC standard unnecessary, which in turn allows its faster implementation on a large scale. The proposed approach is also compared with other synchronized methods presented in recent literature and with methods suggested by IEC standards. The presented research results confirm the accuracy of the proposed solution.

Lithium-ion batteries are widely used in electric vehicles due to their high specific energy value. The continuous charge/discharge cycles pf batteries, determining a change of their State of Health (SoH) and a reduction of their Remaining Useful Life (RUL). The relevant literature describes a variety of methods for the prognostic of battery degradation and RUL estimation, based on electrochemical models or on statistical and artificial intelligence techniques. However, the capacity degradation of lithium-ion batteries depends on many uncertain parameters that are suitable to be estimated and modeled within a probabilistic framework. Therefore, the development of new probabilistic methodologies for the prognostic of lithium-ion batteries can be useful and effective. In this paper, probabilistic models based on time series (AutoRegressive Integrated Moving Average model with eXogenous predictors, (ARIMAX)) and regression approaches (Linear Quantile Regression (LQR), Bootstrap Multiple Linear Regression (B-MLR) and Bayesian Bootstrap Multiple Linear Regression (BB-MLR)) are developed and compared for this purpose. All the considered models are specifically suited up to exploit data coming from Accelerated Degradation Tests (ADTs), and developed under two different approaches (i.e., non-differentiation and differentiation of the target time series) to predict SoH and RUL. Moreover, a dedicated procedure to extract a single, point value from the probabilistic predictions is presented to let the models work also in deterministic scenarios. The performance and the comparison between the proposals and several benchmarks taken from the literature are carried out using data from publicly available databases, confirming the validity and the accuracy of the proposed solutions.

In this paper, we assess the effectiveness of forecasting and optimization algorithms on a laboratory demonstration platform that mimics a domestic distribution grid with a high penetration of photovoltaic (PV) systems. Despite the uncertainties, the considered algorithms ensure efficient and secure real-time (RT) operation of the distribution grid, as well as the provision of flexibility services from the low-voltage (LV) distribution grid to the upstream medium-voltage (MV) grid. Uncertainties arise from the variations in PV systems power production and end-users’ power consumption, as well as RT deployment of flexibility services. As a result of the considered algorithms, the distribution grid becomes active in the provision of flexibility services. The forecasting and optimization algorithms are based on Bayesian bootstrap quantile regression (BBQR) and distributionally robust chance-constrained (DRCC) programming, respectively. This paper also evaluates the framework of the laboratory demonstration platform for the deployment of the considered algorithms.