Search results for "Predictive Model"
showing 10 items of 74 documents
External parameters contribution in domestic load forecasting using neural network
2015
Domestic demand prediction is very important for home energy management system and also for peak reduction in the power system network. In this work, for precise prediction of power demand, external parameters, such as temperature and solar radiation, are considered and included in the prediction model for improving prediction performance. Power prediction models for coming hours' power demand estimation are built using neural network based on hourly power consumptions data with / without ambient temperature data and global solar irradiation (GSI) data respectively. In this work, a typical Southern Norwegian household demand has been considered. As a result, both ambient temperature and GSI…
Introducing the Human Factor in Predictive Modelling: a Work in Progress
2012
International audience; In this paper we present the results of a study into integrating socio-cultural factors into predictive modelling. So far, predictive modelling has largely neglected the social and cultural dimensions of past landscapes. To maintain its value for archaeological research, therefore, it needs new methodologies, concepts and theories. For this study, we have departed from the methodology developed in the 1990s during the Archaeomedes Project. In this project, cross-regional comparisons of settlement location factors were made by analyzing the environmental context of Roman settlements in the French Rhône Valley. For the current research, we expanded the set of variables…
A Comprehensive Check of Usle-Based Soil Loss Prediction Models at the Sparacia (South Italy) Site
2020
At first, in this paper a general definition of the event rainfall-runoff erosivity factor for the USLE-based models, REFe = (QR)b1(EI30)b2, in which QR is the event runoff coefficient, EI30 is the single-storm erosion index and b1 and b2 are coefficients, was introduced. The rainfall-runoff erosivity factors of the USLE (b1 = 0, b2 = 1), USLE-M (b1 = b2 = 1), USLE-MB (b1 ≠ 1, b2 = 1), USLE-MR (b1 = 1, b2 ≠ 1), USLE-MM (b1 = b2 ≠ 1) and USLE-M2 (b1 ≠ b2 ≠ 1) can be defined using REFe. Then, the different expressions of REFe were simultaneously tested against a dataset of normalized bare plot soil losses, AeN, collected at the Sparacia (south Italy) site. As expected, the poorest AeN predict…
OPTIMIZING STOCHASTIC SUSCEPTIBILITY MODELLING FOR DEBRIS FLOW LANDSLIDES: MODEL EXPORTATION, STATISTICAL TECHNIQUES COMPARISON AND USE OF REMOTE SEN…
Il presente lavoro di ricerca è stato sviluppato al fine di approfondire approcci metodologici nell'ambito della sucscettibilità da frana. In particolare, il tema centrale della ricerca è rappresentato dal tema specifico dell'esportazione spaziale di modelli di suscettibilità nell'area mediterranea. All'interno del topic specifico dell'esportazione di modelli predittivi spaziali sono state approfondite tematiche relative all'utilizzo di differenti algoritmi o di differenti sorgenti, derivate da DEM o da coperture satellitari. The present work has been developed in order to enhance current methodological approaches within the big picture of the landslide susceptibility. In particular, the ce…
A distributed real-time data prediction and adaptive sensing approach for wireless sensor networks
2018
International audience; Many approaches have been proposed in the literature to reduce energy consumption in Wireless Sensor Networks (WSNs). Influenced by the fact that radio communication and sensing are considered to be the most energy consuming activities in such networks. Most of these approaches focused on either reducing the number of collected data using adaptive sampling techniques or on reducing the number of data transmitted over the network using prediction models. In this article, we propose a novel prediction-based data reduction method. furthermore, we combine it with an adaptive sampling rate technique, allowing us to significantly decrease energy consumption and extend the …
Predictive role of capillaroscopic skin ulcer risk index in systemic sclerosis: A multicentre validation study
2012
IntroductionThe early detection of systemic sclerosis (SSc) patients at high risk of developing digital ulcers could allow preventive treatment, with a reduction of morbidity and social costs. In 2009, a quantitative score, the capillaroscopic skin ulcer risk index (CSURI), calculated according to the formula ‘D×M/N2’, was proposed, which was highly predictive of the appearance of scleroderma digital ulcers within 3 months of capillaroscopic evaluation.ObjectivesThis multicentre study aims to validate the predictive value and reproducibility of CSURI in a large population of SSc patients.MethodsCSURI was analysed in 229 unselected SSc patients by nailfold videocapillaroscopy (NVC). All pati…
Basis of predictive mycology.
2005
Abstract For over 20 years, predictive microbiology focused on food-pathogenic bacteria. Few studies concerned modelling fungal development. On one hand, most of food mycologists are not familiar with modelling techniques; on the other hand, people involved in modelling are developing tools dedicated to bacteria. Therefore, there is a tendency to extend the use of models that were developed for bacteria to moulds. However, some mould specificities should be taken into account. The use of specific models for predicting germination and growth of fungi was advocated previously [ Dantigny, P., Guilmart, A., Bensoussan, M., 2003. Basis of predictive mycology. In Proceedings of the 4th Internatio…
Comparison of machine learning and logistic regression as predictive models for adverse maternal and neonatal outcomes of preeclampsia: A retrospecti…
2022
IntroductionPreeclampsia, one of the leading causes of maternal and fetal morbidity and mortality, demands accurate predictive models for the lack of effective treatment. Predictive models based on machine learning algorithms demonstrate promising potential, while there is a controversial discussion about whether machine learning methods should be recommended preferably, compared to traditional statistical models.MethodsWe employed both logistic regression and six machine learning methods as binary predictive models for a dataset containing 733 women diagnosed with preeclampsia. Participants were grouped by four different pregnancy outcomes. After the imputation of missing values, statistic…
Dynamic Modeling and Driving Cycle Prediction for a Racing Series Hybrid Car
2014
International audience; This paper presents Noao, a plug-in series hybrid racing car equipped with an engine/generator set as range extender. To determine the velocity profile, i.e., performance of the car and its power profile, a dynamic model for this car is developed using pedal position as input. This value is easy to measure, representative for race cycles, and presents a novelty. The model is validated with the results from experiments. An analysis based on the map of Magny-Cours racing circuit and drivers pedal action on certain zones of the circuit is formulated and is used as a prediction tool to determine drivers inputs on other racing circuits and generate driving schedules. The …