Giuseppe De Nicolao

Alessandro Incremona, Giuseppe De Nicolao

The problem of forecasting the whole 24 profile of the Italian electric load is addressed as a multitask learning problem, whose complexity is kept under control via alternative regularization methods. In view of the quarter-hourly samplings, 96 predictors are used, each of which linearly depends on 96 regressors. The 96x96 matrix weights form a 96x96 matrix, that can be seen and displayed as a surface sampled on a square domain. Different regularization and sparsity approaches to reduce the degrees of freedom of the surface were explored, comparing the obtained forecasts with those of the Italian Transmission System Operator Terna. Besides outperforming Terna in terms of quarter-hourly mean absolute percentage error and mean absolute error, the prediction residuals turned out to be weakly correlated with Terna, which suggests that further improvement could ensue from forecasts aggregation. In fact, the aggregated forecasts yielded further relevant drops in terms of quarter-hourly and daily mean absolute percentage error, mean absolute error and root mean square error (up to 30%) over the three test years considered.

Emanuele Fabbiani, Pulkit Nahata, Giuseppe De Nicolao et al.

The increasing penetration of intermittent distributed energy resources in power networks calls for novel planning and control methodologies which hinge on detailed knowledge of the grid. However, reliable information concerning the system topology and parameters may be missing or outdated for temporally varying electric distribution networks. This paper proposes an online learning procedure to estimate the network admittance matrix capturing topological information and line parameters. We start off by providing a recursive identification algorithm exploiting phasor measurements of voltages and currents. With the goal of accelerating convergence, we subsequently complement our base algorithm with a design-of-experiment procedure which maximizes the information content of data at each step by computing optimal voltage excitations. Our approach improves on existing techniques, and its effectiveness is substantiated by numerical studies on realistic testbeds.

Emanuele Fabbiani, Andrea Marziali, Giuseppe De Nicolao

Gas demand is made of three components: Residential, Industrial, and Thermoelectric Gas Demand. Herein, the one-day-ahead prediction of each component is studied, using Italian data as a case study. Statistical properties and relationships with temperature are discussed, as a preliminary step for an effective feature selection. Nine "base forecasters" are implemented and compared: Ridge Regression, Gaussian Processes, Nearest Neighbours, Artificial Neural Networks, Torus Model, LASSO, Elastic Net, Random Forest, and Support Vector Regression (SVR). Based on them, four ensemble predictors are crafted: simple average, weighted average, subset average, and SVR aggregation. We found that ensemble predictors perform consistently better than base ones. Moreover, our models outperformed Transmission System Operator (TSO) predictions in a two-year out-of-sample validation. Such results suggest that combining predictors may lead to significant performance improvements in gas demand forecasting.

Andrea Marziali, Emanuele Fabbiani, Giuseppe De Nicolao

Gas demand forecasting is a critical task for energy providers as it impacts on pipe reservation and stock planning. In this paper, the one-day-ahead forecasting of residential gas demand at country level is investigated by implementing and comparing five models: Ridge Regression, Gaussian Process (GP), k-Nearest Neighbour, Artificial Neural Network (ANN), and Torus Model. Italian demand data from 2007 to 2017 are used for training and testing the proposed algorithms. The choice of the relevant covariates and the most significant aspects of the pre-processing and feature extraction steps are discussed in-depth, lending particular attention to the role of one-day-ahead temperature forecasts. Our best model, in terms of Root Mean Squared Error (RMSE), is the ANN, closely followed by the GP. If the Mean Absolute Error (MAE) is taken as an error measure, the GP becomes the best model, although by a narrow margin. A main novel contribution is the development of a model describing the propagation of temperature errors to gas forecasting errors that is successfully validated on experimental data. Being able to predict the quantitative impact of temperature forecasts on gas forecasts could be useful in order to assess potential improvement margins associated with more sophisticated weather forecasts. On the Italian data, it is shown that temperature forecast errors account for some 18% of the mean squared error of gas demand forecasts provided by ANN.

Gianluigi Pillonetto, Tianshi Chen, Alessandro Chiuso et al.

Inspired by ideas taken from the machine learning literature, new regularization techniques have been recently introduced in linear system identification. In particular, all the adopted estimators solve a regularized least squares problem, differing in the nature of the penalty term assigned to the impulse response. Popular choices include atomic and nuclear norms (applied to Hankel matrices) as well as norms induced by the so called stable spline kernels. In this paper, a comparative study of estimators based on these different types of regularizers is reported. Our findings reveal that stable spline kernels outperform approaches based on atomic and nuclear norms since they suitably embed information on impulse response stability and smoothness. This point is illustrated using the Bayesian interpretation of regularization. We also design a new class of regularizers defined by "integral" versions of stable spline/TC kernels. Under quite realistic experimental conditions, the new estimators outperform classical prediction error methods also when the latter are equipped with an oracle for model order selection.