Search

International audience

Two types of probabilistic forecasts are used in the field of solar energy forecasting. The first one takes the form of Ensemble forecasts commonly provided by meteorological utilities such as ECMWF. The second one is based on statistical methods that generate quantiles forecasts. The first type of forecasts is treated in the first part of this article. In this second part, we focus on the assessment of the quality of quantile forecasts characterized by two main attributes namely reliability and resolution. More precisely, we propose a verification framework based on diagnostic tools and quantitative scoring rules that are specifically designed for the evaluation of quantile forecasts. The probabilistic models are evaluated on two selected sites that experience very different climatic conditions. It is shown that the decomposition into reliability and resolution of a scoring rule, the Continuous Ranked Probability Score (CRPS), may help to obtain a detailed picture of the performance of the models and consequently may help in selecting the best probabilistic method.

This paper presents the Post Occupancy Evaluation of the first naturally ventilated lecture theatre built in a tropical region. The challenge was to design a bioclimatic amphitheatre of 550 seats, which does not use air conditioning and remains comfortable at the same time. A post-occupancy evaluation has been carried out based on, in parallel, indoor and outdoor measurements, and a user’s survey. Airflow measurement is used to check the good performance in terms of natural ventilation. A follow-up of the building’s actual consumption has been carried out as well, which showed overall positive results. 92% of the users feel comfortable or slightly comfortable during the hottest days. Nevertheless, 50% of them would prefer to get more air movements. In terms of energy consumption, the lecture theatre consumes four times less than a standard one.

International audience; One of the handicaps of the large-scale integration of solar energy is due to its variability and its intermittency. The main way to overcome this issue is the energy storage technology. Knowing the high cost of batteries and their impact on the environment, we simulate a storage system based on compressed air and acting as a battery system. The CAES consists in storing the air at a high pressure in a tank during the period when the energy source is abundant, i.e., cheap, or when the energy demand is low. The compressed air is later expanded through an air turbine which generates electricity during the high demand periods, i.e. when the energy source becomes very expensive for instance. This system could be used for decentralized electricity supply or in an area with no electric grid. In order to evaluate the feasibility of a Compressed Air Energy Storage system coupled to a photovoltaic plant and a building that represents a reduced power demand, a numerical model that reflects the instant behaviour has been built. The system is composed of a photovoltaic power plant, an air compression system, a storage vessel, an expansion module, a power grid and a building. The inputs used are, on the one hand, the climate data such as ambient temperature and the global solar irradiation and, on the other hand, the load curve of a building or of the group of buildings, which has to be supplied by electricity. The overall system optimization has then been performed after having done a sensitivity analysis of the key parameters. This optimization allows us to find the most suitable size for each component of the system: compressor, tank size and photovoltaic area. Abstract One of the handicaps of the large-scale integration of solar energy is due to its variability and its intermittency. The main way to overcome this issue is the energy storage technology. Knowing the high cost of batteries and their impact on the environment, we simulate a storage system based on compressed air and acting as a battery system. The CAES consists in storing the air at a high pressure in a tank during the period when the energy source is abundant, i.e., cheap, or when the energy demand is low. The compressed air is later expanded through an air turbine which generates electricity during the high demand periods, i.e. when the energy source becomes very expensive for instance. This system could be used for decentralized electricity supply or in an area with no electric grid. In order to evaluate the feasibility of a Compressed Air Energy Storage system coupled to a photovoltaic plant and a building that represents a reduced power demand, a numerical model that reflects the instant behaviour has been built. The system is composed of a photovoltaic power plant, an air compression system, a storage vessel, an expansion module, a power grid and a building. The inputs used are, on the one hand, the climate data such as ambient temperature and the global solar irradiation and, on the other hand, the load curve of a building or of the group of buildings, which has to be supplied by electricity. The overall system optimization has then been performed after having done a sensitivity analysis of the key parameters. This optimization allows us to find the most suitable size for each component of the system: compressor, tank size and photovoltaic area.

International audience; In this paper, the geographical dispersion of PV installations and Energy Storage System (ESS) are studied to smooth PV production on Reunion Island. Despite the small area of the Island, the results show that with dispersion at the total grid scale, the variability driven by weather conditions (unpredictable variability) is almost removed. Then a centralized EES management at grid scale and a few large EES units could mitigate the variability even further with a relative low cost.

Wounded soldier pre hospital covering: the forward medicalisation. Forward medicalisation is, for French medical service, the first link of wounded soldier all round chain, whose goal is the more precocity arrival in a military hospital on national field after has been done the necessary measures for the survival and prevention of sequelae during all this chain. This forward medicalisation is based on the experiences of anterior conflicts and modern medical technical evolution. It works around combat casualties care, declined in three competences levels. Under enemy fire, each soldier is capable of implementing first combat casualties care level. A second level corresponds for producing of rescue technical acts by specifically trained personals, until the arrival of a medical staff witch is most of the time a doctor and a nurse. This pair, trained at the third combat casualties care level, is able to apply forward reanimation acts and to prepare the evacuation of the wounded in direction to the first surgical field team.Prise en charge préhospitalière du blessé de guerre : la médicalisation de l’avant. La médicalisation de l’avant correspond pour le service de santé des armées au premier maillon d’une chaîne de prise en charge globale du blessé de guerre, dont l’objectif est son arrivée la plus précoce possible dans une structure hospitalière militaire sur le territoire national, après avoir réalisé les gestes nécessaires à sa survie et à la prévention des séquelles tout au long de cette chaîne. Cette médicalisation de l’avant se fonde sur l’expérience des conflits antérieurs et l’évolution des techniques de soins modernes. Elle s’articule autour du sauvetage au combat, qui se décline en trois niveaux de compétences. Sous le feu ennemi, chaque combattant est capable de mettre en oeuvre le sauvetage au combat de niveau 1. Un deuxième niveau correspond à la réalisation de gestes techniques salvateurs par des personnels spécifiquement formés, jusqu’à l’arrivée d’une équipe médicale comprenant le plus souvent un médecin et un infirmier. Ce binôme, formé au sauvetage au combat de niveau 3, est en mesure d’appliquer des gestes de réanimation de l’avant et de préparer l’évacuation du blessé vers la première structure chirurgicale de terrain.

International audience; Solar forecasting has become an important issue for power systems planning and operating, especially in islands grids. Power generationand grid utilities need day ahead, intra-day and intra-hour Global Horizontal solar Irradiance (GHI) forecasts for operations. Inthis paper, we focus on intra-day solar forecasting with forecast horizons ranging from 1 h to 6 h ahead. An Artificial Neural Networks(ANN) model is proposed to forecast GHI using ground measurement data and satellite data (from Helioclim-3) as inputs. In order tocompare the forecasting results obtained by the proposed ANN model, we also include in this work a simple naı¨ve model, based on thepersistence of the clear sky index (smart persistence model), as well as another reference model, the climatological mean model. The modelswere trained and tested for two ground measurements stations in Gran Canaria Island, Pozo (south) and Las Palmas (in the north).Firstly, ANN was trained and tested only with past ground measurement irradiance and compared by means of relative metrics withnaı¨ve models. While this first step led to better performances, forecasting skills were improved by including exogenous inputs to themodel by using GHI satellite data from surrounding area.

This work compares the performance of machine learning methods (k-nearest-neighbors (kNN) and gradient boosting (GB)) in intra-hour forecasting of global (GHI) and direct normal (DNI) irradiances. The models predict the GHI and DNI and the corresponding prediction intervals. The data used in this work include pyranometer measurements of GHI and DNI and sky images. Point forecasts are evaluated using bulk error metrics while the performance of the probabilistic forecasts are quantified using metrics such as Prediction Interval Coverage Probability (PICP), Prediction Interval Normalized Averaged Width (PINAW) and the Continuous Ranked Probability Score (CRPS). Graphical verification displays like reliability diagram and rank histogram are used to assess the probabilistic forecasts. Results show that the machine learning models achieve significant forecast improvements over the reference model. The reduction in the RMSE translates into forecasting skills ranging between 8% and 24%, and 10% and 30% for the GHI and DNI testing set, respectively. CRPS skill scores of 42% and 62% are obtained respectively for GHI and DNI probabilistic forecasts. Regarding the point forecasts, the GB method performs better than the kNN method when sky image features are included in the model. Conversely, for probabilistic forecasts the kNN exhibits rather good performance.

Accurate solar forecasts are necessary to improve the integration of solar renewables into the energy grid. In recent years, numerous methods have been developed for predicting the solar irradiance or the output of solar renewables. By definition, a forecast is uncertain. Thus, the models developed predict the mean and the associated uncertainty. Comparisons are therefore necessary and useful for assessing the skill and accuracy of these new methods in the field of solar energy. The aim of this paper is to present a comparison of various models that provide probabilistic forecasts of the solar irradiance within a very strict framework. Indeed, we consider focusing on intraday forecasts, with lead times ranging from 1 to 6 h. The models selected use only endogenous inputs for generating the forecasts. In other words, the only inputs of the models are the past solar irradiance data. In this context, the most common way of generating the forecasts is to combine point forecasting methods with probabilistic approaches in order to provide prediction intervals for the solar irradiance forecasts. For this task, we selected from the literature three point forecasting models (recursive autoregressive and moving average (ARMA), coupled autoregressive and dynamical system (CARDS), and neural network (NN)), and seven methods for assessing the distribution of their error (linear model in quantile regression (LMQR), weighted quantile regression (WQR), quantile regression neural network (QRNN), recursive generalized autoregressive conditional heteroskedasticity (GARCHrls), sieve bootstrap (SB), quantile regression forest (QRF), and gradient boosting decision trees (GBDT)), leading to a comparison of 20 combinations of models. None of the model combinations clearly outperform the others; nevertheless, some trends emerge from the comparison. First, the use of the clear sky index ensures the accuracy of the forecasts. This derived parameter permits time series to be deseasonalized with missing data, and is also a good explanatory variable of the distribution of the forecasting errors. Second, regardless of the point forecasting method used, linear models in quantile regression, weighted quantile regression and gradient boosting decision trees are able to forecast the prediction intervals accurately.

In contrast to deterministic forecasts, probabilistic forecasts give additional information about the inherent uncertainty embodied in weather predictions. In the realm of solar forecasting, prediction intervals are especially important to assess risks in grid operations and to optimize the energy storages needed to ensure the supply–demand balance. Even if the development of probabilistic solar forecasts is relatively recent, the main available methods come from other fields of meteorology, particularly from the wind domain. This chapter reviews some of the methods used to generate probabilistic solar forecasts. A special emphasis is put on short term (from several hours to several days) and very short term (from several minutes to several hours) forecasts. As the verification of the quality of the probabilistic forecasts is of major interest, graphical tools like reliability diagram and rank histogram are depicted. These diagnostic tools are relevant for assessing the good calibration of the probabilistic forecasts. In addition, a quantitative score, the CRPS, is also proposed. The CRPS takes into account the different sources of uncertainties and as a proper score, the CRPS is useful to rank competing forecasting methods.

This chapter aims at characterizing and modeling solar resource variability. It is shown that understanding solar energy variability requires a definition of the temporal and spatial context for which variability is assessed. This research describes a predictable, quantifiable variability-smoothing space–time continuum from a single point to thousands of kilometers and from seconds to days. Implications for solar penetration on the power grid and variability mitigation strategies are also discussed. Models for predicting intra-day or intra-hourly variability as a function of insolation conditions are also depicted.