The decentralisation of energy supply, largely driven by renewable energy sources, has led to increased volatility and imbalances in the power grid. To maintain grid stability, flexible balancing mechanisms are required. Ancillary services offer this flexibility by either injecting power (upward regulation) or absorbing it (downward regulation). Among these services, the Automatic Frequency Restoration Reserve (aFRR) plays a crucial role due to its large regulating capacity and direct impact on imbalance prices through its activation costs. As market volatility increases, short-term trading in intra-day and imbalance markets becomes more important, increasing the need for accurate forecasting. While day-ahead and intra-day market forecasting is well-established in Electricity Price Forecasting (EPF), there is limited research on imbalance markets, especially the Dutch aFRR market, highlighting the novelty of this study. The Dutch market uses a dual pricing system with separate bid ladders for upward and downward regulation.

This research focuses on forecasting aFRR bid ladders in the Dutch electricity market and explores their application in short-term trading strategies. Forecasts are generated three hours ahead of delivery, aligning with decision points for intra-day trading, aFRR participation, or opting for no action. The study includes a detailed market analysis, a review of forecasting methods, the development of a machine learning framework tailored to the aFRR market, evaluation of forecast performance, and the practical use of these forecasts in trading scenarios.

A structured machine learning pipeline is designed, encompassing data pre-processing, transformation, model selection, prediction, and evaluation. During the transformation phase, data is scaled and then reduced in dimensionality using Principal Component Analysis (PCA) to retain key variance. The resulting components are used as inputs for predictive models, including LASSO, XGBoost, and LSTM, which are benchmarked against preliminary bid ladders published three hours before delivery. Forecast accuracy is evaluated using point metrics (sMAPE), interval metrics (PICP and PINAW), and a novel self-developed metric called the Largest Knick Volume (LKV), which captures accuracy at key inflection points in the bid ladders that are most relevant for short-term trading.

The findings are twofold. First, model evaluations both at the PCA level and on the reconstructed bid ladders indicate that all models can track general market trends but do not outperform the benchmark. Diebold-Mariano tests confirm that benchmark performance is superior at the PCA level. After reconstructing the bid ladders, benchmark sMAPE scores are 7% for upward and 8% for downward regulation, outperforming LASSO (7%/10%), XGBoost (8%/11%), and LSTM (7%/11%). Second, the forecasts are integrated into Battery Energy Storage System (BESS) intra-day trading strategies, including one based solely on intra-day prices and three incorporating different aFRR bid ladder positions: gas turbine marginal cost, intra-day price with a premium, and LKV-based positioning. Integrating aFRR forecasts improves trading performance, with some high volume-price strategies boosting revenue by up to 18%.

Although the developed forecasting methods do not outperform the benchmark, the benchmark itself already captures much of the relevant market information available three hours ahead, limiting the scope for additional forecasting improvements. This limitation is attributed to the weak correlation between market fundamentals and bid outcomes, as well as the strong influence of individual actors in the relatively small Dutch aFRR market. Nonetheless, the study shows that using existing market data and participating across multiple markets can improve profitability. The results support the strategic value of data-driven, forecast-informed bidding for enhancing the economic performance of battery storage systems.

This study explores how atmospheric temperature affects the rate of muons detected by KM3NeT’s ORCA and ARCA detectors. By comparing measured rates with simulations, a clear seasonal pattern is found, but unexpected differences suggest issues with the simulations. Adjustments to the data improve results but don’t fully resolve the discrepancies, pointing to areas for further investigation.

Given the escalating impacts of climate change, the shift towards more resilient and sustainable energy systems is essential. This higher reliance in renewable energy sources, such as solar power, demands greater accuracy in forecasting changes in weather conditions. This research introduces a novel technique to enhance the accuracy of current solar radiation forecasts on stratocumulus days, by combining satellite observations and large-eddy simulations (LES). Two methods were proposed and implemented in DALES to study the evolution of the cloud field: (i) an advection-only scheme, disregarding all physical processes except horizontal advection, and (ii) standard LES, with application of nudging during the spin-up period of the model.

The implementation of the advection-only model in DALES demonstrated that this approach is not completely successful in isolating the role played by horizontal advection from the remaining physical processes commanding cloud evolution. The rise of non-zero subfilter-scale turbulent fluxes throughout the boundary layer was observed, suggesting diffusion of the thermodynamic fields. In the standard LES runs, the application of nudging (using time scales of 60s and 300s) during the spin-up period of the model fulfilled its intention of keeping a mean thermodynamic state close to the initially prescribed vertical profiles. Nonetheless, this has shown to compromise the development of turbulence in the system, especially in the sub-cloud layer, leading to an underestimation of the turbulent fluxes in the model for this region. Analysis suggested that initialising DALES with vertical profiles for the thermodynamic quantities showing zero mean vertical gradients in the boundary layer are a plausible justification for the results obtained, as this would lead to reduced variances of the thermodynamic quantities and a subsequent underestimation of the turbulent fluxes observed.

Despite the great agreement shown between the vertical profiles coming from ground-based and satellite observations, using the technique proposed in this research, and the radiosonde measurements, one of the main conclusions from this research is the need to continue exploring the application of nudging. This includes performing test simulations with a wide range of parameters commanding nudging (e.g. spin-up duration or nudging time scale), and with the prescription of initial vertical profiles exhibiting some curvature in the boundary layer to allow for an increase in the variances of the thermodynamic quantities in the domain. As findings revealed that a genuine horizontal advection using DALES was not attained, the usage of the current advection-only module in DALES is not recommended, and further research is advised until turbulent effects are completely mitigated.

More and more research shows the substantial health repercussions
of air pollution. Therefore, improving air quality is high on political agendas in modern societies. In the Netherlands, particularly around major roads, NO2 standards set by the government are often exceeded. Air quality models are used to monitor air quality values and design policies to reduce air pollution. Currently, authorities in the Netherlands use a Gaussian Plume Model for decision-making, but this model paints a rather skewed view of reality due to
its underlying assumptions. This research contributes to academic
knowledge about air quality modelling by evaluating two innovative model types, a physics-based LES model and a data-driven regression model, for their usage in decision-making to improve air quality. This is done by comparing the performance of both models with the performance of a Gaussian Plume Model for predicting NO2 levels around a large highway in the Netherlands. Also, two
combinations of the LES model and the regression model are examined. It is concluded that both the LES model and the regression model show potential for accurately predicting air quality around highways in the Netherlands. The LES model is particularly suitable for predicting high NO2 levels, and the regression model is considered suitable for predicting the average NO2 levels over a longer timeframe. A model in which the LES results were combined with a
regression model outperformed the original models and is therefore considered to hold the most potential for usage within air quality
policy.

With the growing global drive to act up on climate change, the adoption of renewable energy sources such as solar photovoltaic (PV) is continuously increasing. This crucial shift poses many economic and environmental benefits, however the variability in solar PV generation may also threaten the stability of our power grid and energy supply. The reliable prediction of this fluctuating power resource on various time scales has been identified as a crucial technology for the continuous massive adoption of solar PV. This study concentrates on the application of convolutional neural networks (CNN) and Long Short Term Memory (LSTM) to process real-time data sources in spatially aggregated solar PV power forecast for Germany, with specifically a forecast horizon of 3 hours and 15-minute interval. Two models are designed to be applicable in a real-time operational setting with a short forecast lag: (1) A LSTM network that leverages on the latest solar PV generation data and a NWP based day-ahead power forecast, and (2) a CNN-LSTM network designed to utilize the latest satellite images and a NWP based day-ahead power forecast. The accuracy of the forecast models are evaluated using one year of solar PV power generation data in Germany (January 2020 through December 2020), and are compared to a persistence model and a NWP based day-ahead and intra-day power forecast provided by the German transmission system operators. The empirical results show that the two proposed models perform equal or better than the benchmark models. An implication for power trading practices is that deep learning models, such as LSTM and CNN-LSTM, shows to be a promising forecasting technique which deserves a place in a comprehensive solar PV power forecasting toolbox.

The goal of this research is to be used as a framework for adding energy storage to aid in the connection of decentralized renewable energy generation in areas with limited connection availability to the electricity grid. The dependence on fossil fuels is reduced with the aim to reduce greenhouse gas emissions. In the Netherlands this means that natural gas is slowly faded out of our society. One consequence of this is the electrification of some functionalities that natural gas had, like heating houses and cooking. Additionally, more and more people are choosing to drive electric vehicles. On the supply side more and more renewable energy sources are being installed to increase the renewable energy mix in our electricity market. Because the supply and demand of electricity increases faster than new cables can be installed, the electricity grid is put under strain. The electricity grid was designed, decades ago, to only transport electricity from central, large energy generation plants to the consumer. However, nowadays there are also developers that are building 'decentralized' PV parks and wind farms. These decentralized generation systems are unable to be connected to the MV-grid. Energy storage has been identified to be a valuable asset to help with the penetration of renewable energy. Energy storage can help to match the supply and demand, improve the power quality, improve the variability of the power supply and smooth out the peaks of over-generation most associated with the congestion on the MV-grid. It is also possible to (temporally) replace the necessary grid reinforcement. However, energy storage is still very expensive. So it needs to be examined how energy storage can be interesting for this problem. A knowledge gap exist on energy storage for decentralized renewable energy generation for the Dutch medium-voltage grid. Additionally, no method was available to simulate such an problem. So this was researched in this study. The study has included a suitable financial model, various energy storage systems and operating strategies. Additionally, the interest of the developer of the DG plant and the DSO were included. For the developer the study has investigated the profitability of the system. The DSO is interested in minimizing the cost and resources of the connection of the hybrid DG system. From this the following research questions are formulated: Under what conditions can energy storage for decentralized renewable energy generation be economically feasible to mitigate connection scarcity on the medium-voltage grid? This was answered in three steps. Firstly, a model was built to accurately simulate energy storage in a hybrid DG system. This model must satisfy all technical requirements and be applicable to the Dutch electricity markets. Secondly, the simulation model was used to examine the hybrid system design to optimally profit from the storage system. From this it was concluded that a small energy storage system adds relatively more value in energy output and earnings than a large storage system. Additionally, it identified the pumped heat electrical storage system as the most profitable system for decentralized generation and the best performing technology is the LFP Li-ion battery. Also, storage is more beneficial for PV solar energy than for wind energy. In this section it was also found that in alleviating connection scarcity, grid reinforcement is more profitable than energy storage. And thirdly, different operating strategies for the hybrid system were investigated to optimally use the energy storage system. For a hybrid system with cable capacity limitations, peak shaving of over-generated energy is the most profitable operating strategy. This research has shown that energy storage can improve the energy output and revenue of a decentralized system with connection capacity limitations. Nevertheless, not adding energy storage is more profitable. A small energy storage system with a PV solar park can become economically feasible and more interesting than grid reinforcement. For this, energy storage technologies need to improve, especially the capital cost need to reduce, with more than half of the current cost. Additionally, this hybrid system needs to be located where grid reinforcement is costly or not possible.

Our present-day society is utterly dependant on electricity. This dependence will only grow as electrification of sectors such as manufacturing, transportation and building heating takes off. Most of the electricity these days comes from conventional plants running on coal and natural gas. Despite that these are reliable and cheap, the disadvantage of emitting greenhouse gases is no longer acceptable. Sustainable alternatives such as solar PV and wind have the highest potential. However, the replacement of conventional power plants by sustainable alternatives is subject to understanding the intermittent and unpredictable behaviour of both wind and solar PV and thereby ensuring that generation equals demand at all times. Energy storage technologies allow for separation between generation and supply to the grid. The aforementioned makes the large-scale integration of wind and solar PV more difficult. This study lays the foundation for an interconnected system model where solar PV, wind, and battery storage is combined. This study is therefore deliberately different than existing studies focussing on small, already severely constrained systems, such as island systems. The problems experienced and possible solutions are first identified by conducting a literature review and simultaneity analysis of wind and solar PV power. It turns out that it is possible while being self-reliant, using all the potential renewable energy and shifting the generation by arbitrage on the APX market to design a system with an increasing rate of income. The simulations showed that when the combined generation (of both wind and solar PV) during peak availability is higher than the grid connection capacity, the computed battery size [MWh] increases rapidly, causing a swift decrease in rate of return. This research also shows that with current imbalance settlement prices and battery installation cost minimizing imbalance is less viable than arbitrage on the APX market. The presented results are consistent with how the electricity system currently operates. At last, the results indicate that curtailing ’cheap’ solar PV energy with significant overplanting on the existing limiting grid connection is beneficial. In this research some important steps have been taken towards the design for a grid-connected optimal system. The method proposed should be tested with more wind and solar PV generation data. Further research should consider longer periods with real generation data making the results presented more accurate.