This study analyzes the production of green hydrogen using dedicated offshore wind power in the Dutch North Sea region. The analysis is based on a detailed techno-economic model that simulates physical flows and estimates the levelized cost of hydrogen (LCOH). However, the model’s outputs depend on user-provided inputs and evaluating all possible inputs is computationally infeasible. To this end, “optimization with constraint learning” is employed, where surrogate machine learning models are trained on simulation data and embedded in mixed-integer optimization problems. The surrogate models are trained on 4096 simulation runs and achieve a mean absolute percentage error of ≤[jls-end-space/]3% for physical flow-related outputs, and an error of ≈[jls-end-space/]10% for the LCOH-related outputs. Once trained, these surrogates enable one to solve stakeholder–specific problem instances in sub-second solve times, supporting rapid scenario analysis and trade-off exploration.

Indirect Demand Response (IDR) programs that incentivize consumer participation through unidirectional price-based mechanisms offer a promising way to mobilize small-scale flexibility. The main challenge in such IDR programs lies in modeling the uncertain price-response relationship of demand response resources (DRRs), which complicates DR pricing. Inverse Optimization (IO) provides an effective method for capturing historical price-response patterns with interpretability and seamless integration into retailers' decision-making frameworks. While traditional IO prioritizes forecast accuracy, recent research advocates for value-oriented forecasting, which prioritizes decision quality to mitigate the practical impacts of forecast errors. Despite their advantages, fully integrated value-oriented approaches can become computationally intensive for IO. To address this, we propose a novel value-oriented IO (VOIO) framework that facilitates a data-driven, value-oriented identification of DRRs' aggregate price-response parameters by shifting IO hyperparameter evaluation metrics to decision-making regret minimization. The proposed method was applied in a case study examining a Balance Responsible Partyś (BRP) participation in the Belgian single-price imbalance market. The results demonstrated that the VOIO approach was able to reduce overall financial losses from forecast errors compared to a forecast-oriented IO benchmark, achieving overall profit gains in the validation and test sets through more conservative pricing.

In Europe, balance responsible parties can deliberately take out-of-balance positions to support transmission system operators (TSOs) in maintaining grid stability and earn profit, a practice called implicit balancing. Model predictive control (MPC) is widely adopted as an effective approach for implicit balancing. The balancing market model accuracy in MPC is critical to decision quality. Previous studies modeled this market using either (i) a convex market clearing approximation, ignoring proactive manual actions by TSOs and the market sub-quarter-hour dynamics, or (ii) machine learning methods, which cannot be directly integrated into MPC. To address these shortcomings, we propose a data-driven balancing market model integrated into MPC using an input convex neural network to ensure convexity while capturing uncertainties. To keep the core network computationally efficient, we incorporate attention-based input gating mechanisms to remove irrelevant data. Evaluating on Belgian data shows that the proposed model both improves MPC decisions and reduces computational time.

Hydrogen and derived fuels may act as long-term energy storage in climate-neutral energy systems. However, risk-averse investors will not invest in sufficient renewable electricity, back-up, electrolyzer and storage capacity if they are only remunerated for the hydrogen or electricity produced and markets for risk are missing. We develop a stochastic equilibrium model to study whether capacity markets can limit costs to consumers by restoring investments risk-neutral levels. Our results show that the efficacy of capacity markets depends on complementary instruments to ensure the availability of renewables. If risk-aversion and missing markets for risk reduce renewable build-out, capacity markets in the electricity and hydrogen sectors are needed to restore the overall capacity mix and limit costs for consumers. If complementary instruments lift investments in renewables, a capacity market in the electricity sector suffices. In this situation, an additional capacity market in the hydrogen sector triggers a bias toward hydrogen-fired backup capacity. This illustrates that an integrated systems perspective is required to design future energy markets.

We assess the effect of different incentive schemes with future financial benefits on photovoltaic (PV) adoption patterns in the residential sector for the two biggest regions of Belgium - Flanders and Wallonia. Due to variation in incentive schemes across the regions and within regions across months, we can identify the effect of changes in future benefits on adoption. We combine and compare three distinct and widely used incentive schemes for PV, namely output-based, capacity-based and net metering, which is electricity cost saving-based. We find that PV adoption in the residential sector is highly sensitive to overall future financial benefits. A one percent increase in overall benefits per kW of installed capacity increases the number of installations by around 6%. However, the different incentive schemes have varying degrees of effectiveness. In particular, capacity- and output-based incentive schemes, which directly compensate households with a fixed price per produced amount of electric energy or per installed capacity annually are around 6 percentage points more effective in fostering PV uptake than the cost saving-based net metering, which depends on future electricity prices and production. Furthermore, the average size of new installations depends on whether incentive schemes remunerate for additional capacity.

Introducing electricity purchase conditions in renewable fuel regulations and carbon accounting is a controversial issue in the US and Europe. We argue that their impact must be assessed considering demand flexibility, local grid conditions, and overlapping policy instruments such as emissions trading schemes and renewable portfolio standards. The introduction of more stringent requirements has a significant impact on companies' reported progress in reducing indirect emissions. In addition, we currently lack reliable carbon intensity data with sufficient temporal and spatial granularity outside the US and Europe. Assessing hydrogen imports should consider strategic behavior. In summary, a broader systems perspective is needed to analyze the impact of electricity purchase conditions on markets, prices, and emissions, integrating voluntary actions to decarbonize hard-to-abate sectors under increasing renewable energy shares. This will have implications not only for the (re)design and integration of certificate markets for renewable electricity but also for renewable fuels and carbon.

Ambitious offshore wind energy targets in the North Sea necessitate innovative solutions for efficiently delivering energy to onshore demand locations. Wind-to-hydrogen systems offer a promising pathway, with three archetypes of system configurations: centralized onshore electrolysis (C-ON), centralized offshore electrolysis (C-OFF), and decentralized offshore electrolysis at each wind turbine (D-OFF). This study introduces a high-resolution, time-dependent simulation framework capable of analyzing offshore wind-to-hydrogen systems with a focus on operational dynamics and comprehensive cost estimation. The framework enables detailed analysis of D-OFF, capturing its unique dynamics driven by direct connections to individual wind turbines, including the impacts of dynamic operation. A comprehensive system analysis, spanning from the wind farm to the hydrogen offtaker, reveals a wide cost range, with Levelized Cost of Hydrogen (LCOHs) ranging from 3.0 to 10.5€/kgH₂ post 2030. Among the different scenarios analyzed, C-OFF with proton exchange membrane electrolysis achieves the lowest LCOHs due to a reduced need for offshore electrical infrastructure, economies of scale, and efficient dynamic operating characteristics. D-OFF with alkaline electrolysis incurs the highest costs and faces operational challenges, such as electrolyzers shutting down when they occasionally fail to reach the minimum load thresholds, lowering hydrogen production. We illustrate the trade-offs between system configurations’ cost, production rate, and electrolyzer stack lifetime across configurations. Insights from this study can be utilized as a starting point for informed decision-making for large-scale wind-to-hydrogen deployment in the Dutch North Sea region.

Demand response is a concept that has been around since the very first electric power systems. However, we have seen an explosion of research on demand response and demand-side technologies in the past 30 years, coinciding with the shift towards liberalized/deregulated electricity markets and efforts to decarbonize the power sector. Now we are also seeing a shift towards more distributed/decentralized electric systems; we have entered the era of “distributed energy resources”, which require new grid management, operational, and control strategies. Given this paradigm shift, we argue that the concept of demand response needs to be revisited, and more carefully/consistently defined to enable us to better utilize this massive resource for economic, technical, environmental, and societal aims.

This paper explores various policies to support climate-neutral hydrogen production, focusing on their interaction with energy markets and cap-and-trade systems such as the EU emission trading scheme. We develop and deploy a state-of-the-art equilibrium model to examine the effect of hydrogen support policies on the interactions between hydrogen, electricity and emission markets. Our analysis shows that mechanisms remunerating hydrogen production can distort spot prices of electricity and hydrogen more strongly than mechanisms that remunerate hydrogen production capacity. Hydrogen support mechanisms furthermore promote renewable electricity production and deter investment in conventional generation assets. The associated decrease in emissions in the power sector leads to an increase of emissions in the industrial and hydrogen sector due to the waterbed effect in the EU emission trading scheme. Our case study on an emission-capped area inspired by the EU shows that the operational distortions that production-based mechanisms exhibit, typically increase costs more than the investment distortions that capacity-based mechanisms entail.

Motivated by generation system adequacy concerns, many European countries have introduced capacity remuneration mechanisms (CRMs) to ensure sufficient investments in power generation. However, it is uncertain whether the existing CRMs will promote sufficient adequacy and flexibility in a decarbonized power system, where supply and demand will become more weather-dependent. We assess the effectiveness of a centralized capacity market, a strategic reserve, and a decentralized capacity market via capacity subscriptions in a climate-neutral, weather-driven power system. We develop a co-simulation of two agent-based models simulating myopia in both operational and investment decisions. We simulate weather uncertainty by running the model with 40 different weather years. Our results from a case study based on the Netherlands indicate that a strategic reserve may increase electricity price volatility in the long-term. A centralized capacity market is more cost-effective than a strategic reserve, but administratively setting its parameters is prone to over- or underprocurement. Capacity subscription allows consumers to select their desired level of reliability. Results indicate that these decentralized capacity markets may yield a clearer signal for the needed dispatchable capacity and promote demand-side response, but it may be challenging to provide long-term certainty for investors.

Due to computational limits, temporal details within Energy System Optimisation Models are often reduced, for example by reducing the time horizon or by resampling via Time Series Aggregation (TSA) techniques. In high RES energy systems, this may lead to undersizing of Long-Duration Energy Storage (LDES) capacities, necessary for system flexibility, due to the omission of long-term interannual weather effects. Via comparative analysis between the capacity expansion results for different subsets of weather years, this paper shows the extent to which single year models underpredict LDES. but also that a small cluster (n=2, 3) of weather years can adequately capture key system-defining weather patterns. Identifying these weather years ex-ante is non-trivial, as there is no obvious correlation with how well they describe the full set of weather years. As this assumed correlation underpins current time series aggregation techniques, new techniques are required.

Increasing shares of renewable generation are leading to more volatile electricity prices, presenting an opportunity for Energy Storage Systems (ESS) participating in short-term electricity markets. Model Predictive Control (MPC) has been shown to be a powerful tool to leverage the latest information at the time of optimization, yet its efficacy depends on the quality of the employed price forecasts. So far, these forecasts have been developed with traditional forecasting methods instead of value-oriented approaches, which consider the downstream decision problem during the forecaster training phase. Existing value-oriented methods, however, often rely on a specific downstream problem structure. This paper addresses these shortcomings by introducing a universally applicable, value-oriented forecasting methodology that employs a generalized loss function designed to account for inter-temporal price variability, using the downstream value (i.e., profit from ESS market participation) as the selection criterion in the hyperparameter tuning step. The proposed methodology is tested on a case study considering different types of ESS participating in the Belgian balancing market through MPC. The method is benchmarked against other forecasting techniques including a neural network trained in traditional, accuracy-oriented fashion. Using real-life data over a test set of two months, we show that the methodology outperforms those traditional techniques in terms of ex-post out-of-sample profit.

Energy Storage Systems (ESS) play a crucial role in managing renewable energy variability. Forecast-informed optimization is typically used to maximize ESS profit in electricity markets. Whereas traditional forecaster training methods use accuracy-based loss functions, Decision-Focused Learning uses a task-aware loss function with the aim of improving ESS profits. This can be achieved by integrating the downstream optimization in the forecaster training procedure. A common task-aware loss function is the Smart Predict-then-Optimize (SPO+) loss. However, its current implementation is prone to overfitting and is limited to linear forecasting models. Here, we extend the SPO+ framework to neural network forecasters with non-linear activation functions while introducing an interior-point training method to mitigate overfitting risks. When applied to an ESS participating in the day-ahead market, our approach outperforms both traditional and other decision-focused benchmarks in terms of obtained ESS profits.

The rapid decarbonisation of the power sector is challenging the traditional resource adequacy framework. Variable and energy-limited resources are driving the emergence of new correlations that, together with extreme weather events, are rapidly changing the expected scarcity conditions in the electricity system. Traditional resource adequacy metrics are showing their limitations under these new conditions, and many regulators have already started to reform them. This article presents the first model-based comparative analysis of two different approaches that have been proposed to overcome these limitations, i.e., multi-metric standards (imposing a set of different resource adequacy constraints) and composite-metric standards (combining different resource adequacy metrics through weighting factors to build a single reliability standard). These two approaches are quantitatively evaluated in this article through case studies obtained from a simulation model, focusing not only on the impact of the reliability standard on the resource mix, but also on the design of the reliability product to be traded in a capacity mechanism to guide the system towards that mix.

The main aim of this paper is to illustrate the added value at system level of explicitly accounting for the closed-loop feedback aspect in the open-loop optimal control problem of model predictive control strategies for thermostatically controlled loads participating in demand response programs, when subject to uncertainties. To this end, an integrated system-level optimization problem is set up, merging an economic dispatch problem to represent the supply side and an open-loop stochastic optimal control problem incorporating affine disturbance feedback to represent the demand side. The incorporation of affine disturbance feedback enables the simultaneous optimal scheduling of the demand for electrical energy, reserve capacity and real-time flexibility required to guarantee thermal comfort at the demand side, thereby disclosing very valuable information for an aggregator or system operator, since the load uncertainty can be revealed and controlled ahead of real time. To solve the mathematically complex integrated problem, a distributed solution strategy based on the alternating direction method of multipliers is developed. With the help of an illustrative case study, it is demonstrated that the day-ahead coordination of the demand for reserve capacity in addition to the energy demand is able to reduce the system operating cost while guaranteeing thermal comfort, and hence, enables a more cost-efficient electrification of the residential heating sector. Cost reductions up to 10.7 % are shown to be achievable for a demand side consisting of 900 000 flexible heat pumps combined with low-temperature radiators.