arXiv subjects

Philippe Lauret

Publications and source records attributed to Philippe Lauret.

4 recordsLinked to original sources

arXiv2025

NICE^k Metrics: Unified and Multidimensional Framework for Evaluating Deterministic Solar Forecasting Accuracy

Accurate solar energy output prediction is key for integrating renewables into grids, maintaining stability, and improving energy management. However, standard error metrics such as Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), and Skill Scores (SS) fail to capture the multidimensional nature of solar irradiance forecasting. These metrics lack sensitivity to forecastability, rely on arbitrary baselines (e.g., clear-sky models), and are poorly suited for operational use. To address this, we introduce the NICEk framework (Normalized Informed Comparison of Errors, with k = 1, 2, 3, Sigma), offering a robust and interpretable evaluation of forecasting models. Each NICEk score corresponds to an Lk norm: NICE1 targets average errors, NICE2 emphasizes large deviations, NICE3 highlights outliers, and NICESigma combines all. Using Monte Carlo simulations and data from 68 stations in the Spanish SIAR network, we evaluated methods including autoregressive models, extreme learning, and smart persistence. Theoretical and empirical results align when assumptions hold (e.g., R^2 ~ 1.0 for NICE2). Most importantly, NICESigma consistently shows higher discriminative power (p < 0.05), outperforming traditional metrics (p > 0.05). The NICEk metrics exhibit stronger statistical significance (e.g., p-values from 10^-6 to 0.004 across horizons) and greater generalizability. They offer a unified and operational alternative to standard error metrics in deterministic solar forecasting.

physics.ao-ph↗

arXiv2024

Coupling Machine Learning Local Predictions with a Computational Fluid Dynamics Solver to Accelerate Transient Buoyant Plume Simulations

Data-driven methods demonstrate considerable potential for accelerating the inherently expensive computational fluid dynamics (CFD) solvers. Nevertheless, pure machine-learning surrogate models face challenges in ensuring physical consistency and scaling up to address real-world problems. This study presents a versatile and scalable hybrid methodology, combining CFD and machine learning, to accelerate long-term incompressible fluid flow simulations without compromising accuracy. A neural network was trained offline using simulated data of various two-dimensional transient buoyant plume flows. The objective was to leverage local features to predict the temporal changes in the pressure field in comparable scenarios. Due to cell-level predictions, the methodology was successfully applied to diverse geometries without additional training. Pressure estimates were employed as initial values to accelerate the pressure-velocity coupling procedure. The results demonstrated an average improvement of 94% in the initial guess for solving the Poisson equation. The first pressure corrector acceleration reached a mean factor of 3, depending on the iterative solver employed. Our work reveals that machine learning estimates at the cell level can enhance the efficiency of CFD iterative linear solvers while maintaining accuracy. Although the scalability of the methodology to more complex cases has yet to be demonstrated, this study underscores the prospective value of domain-specific hybrid solvers for CFD.

physics.flu-dyn↗

arXiv2022

Complex-Valued Time Series Based Solar Irradiance Forecast

This paper describes a new way to predict real time series using complex-valued elements. An example is given in the case of the short-term probabilistic global solar irradiance forecasts with measurement as real part and an estimate of the volatility as imaginary part. A simple complex autoregressive model is tested with data collected in Corsica island (France). Results show that, even if this approach is simple to set up and requires very little resource and data, both deterministic and probabilistic forecasts generated by this model are in agreement with experimental data (root mean square error ranging from 0.196 to 0.325 considering all studied horizons). In addition, it exhibits sometimes a better accuracy than classical models like Gaussian process, bootstrap methodology or even more sophisticated model like quantile regression. The number of models that it is possible to build by generating complex-valued time series is substantial. Indeed, by using exogenous or ordinal variables and computed quantities coupled with complex (or multi-complex) numbers, many studies and many fields of physics could benefit from this methodology and from the many models that result from it.

physics.data-an↗

arXiv2012

A Bayesian Model Committee Approach to Forecasting Global Solar Radiation

This paper proposes to use a rather new modelling approach in the realm of solar radiation forecasting. In this work, two forecasting models: Autoregressive Moving Average (ARMA) and Neural Network (NN) models are combined to form a model committee. The Bayesian inference is used to affect a probability to each model in the committee. Hence, each model's predictions are weighted by their respective probability. The models are fitted to one year of hourly Global Horizontal Irradiance (GHI) measurements. Another year (the test set) is used for making genuine one hour ahead (h+1) out-of-sample forecast comparisons. The proposed approach is benchmarked against the persistence model. The very first results show an improvement brought by this approach.

stat.AP↗