- Energy Load and Power Forecasting
- Building Energy and Comfort Optimization
- Meteorological Phenomena and Simulations
- Diabetes Management and Research
- Forecasting Techniques and Applications
- Solar Radiation and Photovoltaics
- Wastewater Treatment and Nitrogen Removal
- Water resources management and optimization
- Urban Heat Island Mitigation
- Atmospheric and Environmental Gas Dynamics
- Smart Grid Energy Management
- Wind and Air Flow Studies
- Reservoir Engineering and Simulation Methods
- Cosmology and Gravitation Theories
- Diabetes Treatment and Management
- Fault Detection and Control Systems
- Neural Networks and Applications
- Computational Physics and Python Applications
- Advanced Control Systems Optimization
- Particle physics theoretical and experimental studies
- Advanced Battery Technologies Research
- Climate variability and models
- Pancreatic function and diabetes
- Heat Transfer and Optimization
- Water Systems and Optimization
Technical University of Denmark
2014-2024
Dynamic Systems (United States)
2018-2021
Christie's
2019
Capital Region of Denmark
2019
Compute Canada
2018
Bielefeld University
2010-2016
Universität Hamburg
2016
Chalmers University of Technology
2014
Bremen Institute for Applied Beam Technology
2013
Institute of Mathematical Statistics
2006-2011
Abstract Predictions of wind power production for horizons up to 48–72 h ahead comprise a highly valuable input the methods daily management or trading generation. Today, users predictions are not only provided with point predictions, which estimates conditional expectation generation each look‐ahead time, but also uncertainty given by probabilistic forecasts. In order avoid assumptions on shape predictive distributions, these produced from non‐parametric methods, and then take form single...
Reliable probabilistic production forecasts are required to better manage the uncertainty that rapid build-out of wind power capacity adds future energy systems. In this article, we consider sequential methods correct errors in forecast ensembles derived from numerical weather predictions. We propose combining neural networks with time-adaptive quantile regression enhance accuracy forecasts. refer approach as Neural Adaptive Basis for (time-adaptive) Quantile Regression or NABQR. First, use...
Probabilistic forecasts of renewable energy production provide users with valuable information about the uncertainty associated expected generation. Current state‐of‐the‐art for solar irradiance have focused on producing reliable point forecasts. The additional included in probabilistic may be paramount decision makers to efficiently make use this uncertain and variable In paper, a stochastic differential equation framework modeling forecast is proposed. This approach allows characterizing...
Abstract The intermittency of renewable energy sources, such as wind and solar, means that they require reliable accurate forecasts to integrate properly into systems. This review introduces examines a selection state‐of‐the‐art methods are applied for multivariate forecasting solar power production. Methods conditional parametric combined already see wide use in practice, both commercially scientifically. In the context forecasting, it is vital model dependence between correctly. recent...
It has been predicted that electric vehicles will play a crucial role in incorporating large renewable component the energy sector. If are integrated naive way, they may exacerbate issues related to peak demand and transmission capacity limits while not reducing polluting emissions. Optimizing charging of is paramount for their successful integration. This paper presents model describe driving patterns order provide primary input information any mathematical programming optimal charging....
Probabilistic forecasting is becoming increasingly important for a wide range of applications, especially energy systems such as wind power production. A need proper evaluation probabilistic forecasts follows naturally with this, because the key to improving forecasts. Although plenty excellent reviews and research papers on forecast already exist, we find that there an introduction some practical application. In particular, many scenarios in are inherently multivariate, while univariate...
Home Energy Management Systems (HEMSs) are expected to become an inevitable part of the future smart grid technologies. To work effectively, HEMSs require reliable and accurate load forecasts. In this paper, two new modelling methods presented. They both suited for producing multivariate probabilistic forecasts, which consider temporal correlation between forecast horizons. The first method employs point forecasts generated with Recursive Least Squares (RLS) models subsequently analyses...
The increasing penetration of wind power has resulted in larger shares volatile sources supply systems worldwide. In order to operate such efficiently, methods for reliable probabilistic forecasts future production are essential. It is well known that the conditional density highly dependent on level predicted and prediction horizon. This paper describes a new approach forecasting based logistic‐type stochastic differential equations (SDEs). SDE formulation allows us calculate both...
Heat load forecasts are crucial for energy operators in order to optimize the production at district heating plants coming hours. Furthermore, of heat needed optimized control network since a lower temperature reduces loss, but required supply end-users puts limit on level. Consequently, improving accuracy leads savings and reduced loss by enabling improved an schedule plant. This paper proposes use temporal hierarchies enhance heating. Usually, only made aggregation level that is most...
Underestimation of glucose turnover rates has been a problem in clamp studies using primed‐constant [3‐ 3 H]‐glucose infusion technique. Due to slow mixing interstitial compartments concealed specific activity gradients may arise between plasma and during intravenous unlabelled infusion. Such gradients, however, can be prevented if is maintained constant. Two euglycaemic (insulin 40 mU m −2 min −1 ) were performed six lean normal subjects. Using conventional infusates declined by 74%, tracer...
Electric Vehicles (EVs) are considered as one of the important components future intelligent grids. Their role energy storages in electricity grid could provide local sustainable solutions to support more renewable energy. In order estimate extent interaction EVs operation, a careful examination system is essential. This paper investigates degree EV penetration and its key influence on low voltage distribution Three detailed models residential grids Denmark test cases this study, where...
Summary We consider reconciliation of wind power forecasts in a spatial hierarchy with three aggregation levels. produce base for the bottom level consisting 407 substations (connection points local groups turbines). State‐of‐the‐art from commercial forecast provider are available middle and top levels, which consist 15 regions entire Western Denmark (DK1), respectively. find that accuracy total can be improved through reconciliation, even relatively simple model used at lowest hierarchy....
We introduce the open-source Python package NABQR: Neural Adaptive Basis for (time-adaptive) Quantile Regression that provides reliable probabilistic forecasts. NABQR corrects ensembles (scenarios) with LSTM networks and then applies time-adaptive quantile regression to corrected obtain improved more With suggested package, accuracy improvements of up 40% in mean absolute terms can be achieved day-ahead forecasting onshore offshore wind power production Denmark.
Generating flow forecasts with uncertainty limits from rain gauge inputs in sewer systems require simple models identifiable parameters that can adequately describe the stochastic phenomena of system. In this paper, a grey‐box model is proposed attractive for both forecasting and control purposes. The based on differential equations key feature separation total noise into process measurement noise. approach properly introduced hypothesis regarding terms are formulated. Three different...