Abstract

The power grid is the central hub of modern energy systems. Traditional security analysis methods struggle to cope with the rapid changes in operating states arising from high renewable energy penetration. Enhancing real-time security assessment capability is therefore of great importance. This paper proposes a graph neural network (GNN)-based framework for power grid security analysis, which fully exploits the inherent graph structure of power systems and incorporates operating state information to directly learn security-related patterns, enabling fast assessment of grid security. Furthermore, gradient-based attribution is applied to the trained GNN to help quantify and highlight critical buses, transmission lines, and power flow features that influence grid security. Case studies demonstrate that the proposed method can significantly reduce computational cost compared with conventional simulation-based analysis while accurately identifying key security characteristics under different operating conditions.

Abstract

Electricity price forecasting is highly challenging due to market volatility, abrupt spikes, and recurrent seasonal patterns. This study proposes an adaptive and interpretable deep learning framework, termed Sliding-Update Bidirectional LSTM with Attention (SU-BiLSTM-AM), which combines a dynamic update mechanism, bidirectional sequence modeling, and attention-based feature weighting to improve accuracy and robustness under volatile market conditions. Using day-ahead prices from the European Energy Exchange (EEX), SU-BiLSTM-AM consistently outperforms recurrent and sequence-learning baselines, reducing forecast errors by 14-23% compared with LSTM. The model demonstrates robustness under extreme regimes and maintains reasonable computational cost, making it suitable for intelligent energy systems, bidding strategies, and risk management in volatile electricity markets.

Abstract

Oil shale has abundant reserves and is an important alternative resource for traditional oil and natural gas. In situ mining is the most promising method for oil shale, which mainly includes electric heating and steam injection heating methods. However, both methods have obvious drawbacks. Electric heating methods typically require huge amounts of electricity consumption, while steam injection results in significant heat loss from the surface to the ground. This article combines two methods organically and uses wind and solar complementary power to drive an electric heater, thereby heating steam. By introducing the volatility index complementarity rate, the supplementary energy effect of electric heating wells in the all-weather wind solar complementary process can be optimized. Then it is substituted into the in-situ transformation model of oil shale. The results show that during the process of simultaneous electric heating and steam injection, the complementary fluctuation of wind-solar energy supply will generate additional secondary seepage channels. This is mainly due to the instantaneous temperature change, which leads to local thermal stress concentration effect. From the perspective of temperature field, fluctuating temperature changes lead to periodic rapid heating in local areas and also trigger the “thermal stress fatigue” effect. At this point, the permeability mainly increases significantly along the path of the main damage fracture evolution. This provides a new technological approach for efficient in-situ upgrading and development of oil shale.

Abstract

The development of small-scale modular gasification systems represents one of the several pathways needed to follow to reach a net-zero emission society. Currently, the research is fostering the technology readiness level of those units to achieve clean and sustainable energy production from biomass waste while decreasing the load to the centralized waste treatment facilities. The operability of those conversion systems can be improved with the co-gasification of low-grade residues, like sewage sludge, organic municipal solid waste, nut shells, in wood matrixes.
Downdraft co-gasification is a promising technology that offers competitive conversion efficiencies, reliability, and capillary diffusion especially in the context of renewable energy communities. Still, there are research gaps that must be addressed especially for improving the design and process control.
This work presents a modelling approach that allow for deeper understanding of the treatment and propose improvements for SUPREMAS project. A lumped, steady-state model is developed in Aspen Plus using chemical pseudo-equilibrium and empirical correlations, to replicate the real process. Then, the cleaned syngas is converted to energy with a combined heat and power / solid oxide fuel cell units.

Abstract

Buildings in warm-humid climates exhibit pronounced midday cooling peaks, which makes demand-side flexibility under time-of-use tariffs especially valuable. Phase change materials configured in the envelope of walls and roofs can provide latent storage, yet performance under realistic tariffs with explicit hysteresis remains under-quantified. This study develops a reproducible, tariff-linked workflow that couples EnergyPlus with a hysteresis-consistent PCM model and automates a jEPlus parametric sweep. A medium-office archetype in Hong Kong is simulated under a representative four-block tariff, testing overnight pre-cooling to 22 to 24°C and midday set-point relaxation to 24.5 to 25°C, with the PCM melting range aligned to the cooling band. Performance is evaluated using the flexibility factor, on-peak and off-peak energy shifting, peak load moderation and tariff-based cost. The combined strategy delivered a 48.7% daily cooling-cost saving relative to the non-PCM baseline, with on-peak cooling energy reduced by 34.3%, off-peak use increased by 131.1%, and on-peak peak cooling load and peak electric demand reduced by 21.3% and 19.7%, respectively. Single-lever strategies yielded smaller gains, and neglecting hysteresis distorted the timing and magnitude of peaks. The results show that material properties and control could be co-designed, and they provide a transferable workflow with decision-ready indicators for PCM-enabled, tariff-responsive design in warm-humid offices.

Abstract

As the share of renewable energy increases, demand-side flexibility becomes essential. Integrating energy considerations into Production Planning and Control (PPC) systems offers great potential, yet most existing PPC solutions lack this capability. To address this gap, we adopt the approach to treat Energy Flexibility (EF) as an add-on to PPC systems, enhancing rather than extensively modifying or replacing established scheduling processes. We refer to this approach as Energy-flexible Production Rescheduling (EFPR). This paper presents a service architecture for EFPR that supports selected EF measures and objectives. It defines the requirements, describes the needed services and modules as well as PPC system extensions and outlines their interconnections. The Energy Flexibility Data Model (EFDM) is used to standardize the description and exchange of EF information, enabling a factory-wide EF management. The architecture is transferred to a research factory, seamlessly integrating with both the production line and a commercial PPC system, demonstrating the practical applicability and potential of EFPR.

Abstract

The techno-economics of firm block power supply from hybrid solar photovoltaic–battery energy storage systems (BESSs) are central to achieving reliable and dispatchable renewable energy. This study evaluates grid-connected solar-BESS hybrid systems designed to deliver firm 250 MW power blocks over 4-hour, 6-hour, and 8-hour durations during daily morning and evening peaks. Multiple combinations of solar and BESS capacities were analyzed to identify system configurations that minimize the levelized cost of energy (LCOE) while maintaining supply reliability. The BESS was assumed to be charged solely by solar generation, with any surplus solar photovoltaic output curtailed after storage saturation. Results demonstrate that LCOE and reliability are strongly influenced by the solar-to-storage ratio and dispatch duration. Moderate solar oversizing enhances reliability and reduces unmet demand, whereas excessive scaling leads to diminishing economic returns due to higher capital costs and energy curtailment. The study underscores the importance of optimally balancing solar photovoltaic capacity and storage sizing to design cost-efficient, firm renewable-plus-storage systems capable of meeting peak-period demand reliably.

Abstract

Industrial energy supply systems are key enablers for energy-efficient and cost-effective production, yet their operation is characterized by complex dynamics, competing objectives, and increasing uncertainty. While deep reinforcement learning (DRL) has demonstrated significant potential for optimizing control strategies in such systems, its practical adoption remains limited due to the lack of interpretability and transparency of learned policies.
This paper presents an explainability-driven analysis of a DRL-based operational strategy applied to a real-world industrial central cooling system in the metal processing industry. A combination of qualitative behavioral analysis and SHapley Additive exPlanations (SHAP) is used to interpret the learned control policy and to reveal the decision logic underlying the agent’s actions. The DRL-based strategy achieves energy cost reductions of 29 % during training and 15.7 % on an unseen validation year compared to conventional rule-based control.
The explainability analysis shows that the agent autonomously prioritizes free cooling whenever favorable thermal conditions are present and activates mechanical chilling only when internal temperature gradients indicate insufficient heat rejection capability. Furthermore, the results reveal that internal thermodynamic states dominate the agent’s decision-making, while external signals such as electricity price and weather conditions have only minor influence in the investigated system.
By demonstrating how DRL-based control strategies can be interpreted and validated using XAI methods, this work contributes to improving transparency, trust, and acceptance of learning-based control in industrial energy supply systems.

Abstract

Intelligent segmented water control technology is crucial for bottom water reservoir horizontal wells, with control parameters determining application effectiveness. This paper proposes a sustainability oriented optimization method integrating Particle Swarm Optimization (PSO) with reservoir simulation. A unified objective function embedding effective production period as an amplification factor was established, with control timing (τ) and magnitude (α) as decision variables. Three schemes were compared: base case, conventional optimization (maximizing Nₚ), and sustainability oriented optimization (maximizing F with β=0.5). Case study shows the sustainability approach achieved optimal parameters τ*=52.9% and α*=56.8%, yielding cumulative production of 44.40×10⁴ m³ and extending effective production period to 1.80 years—a 28.6% improvement over base case despite only 2.2% production sacrifice compared to conventional optimization (45.38×10⁴ m³, 1.47 years). The proposed method demonstrates that early intervention (τ=50–60%) proves more effective than late-stage control (τ>75%), reducing high-permeability segment contribution from 39.5% to 27.0% and achieving balanced reservoir exploitation. This framework provides quantitative decision support for sustainable horizontal well development in bottom water reservoirs.

Abstract

Isolated islands have a high potential for renewable energy exploitation due to their unique geography. However, the intermittency of these resources requires energy storage systems to ensure a reliable power supply. This study proposes an advanced compressed CO₂ energy storage system integrating a high-temperature thermal energy storage (TES) architecture for isolated island applications. The system exploits seawater static pressure and ocean thermal energy to enable liquid CO₂ isobaric storage, while employing pressurized water and molten salt in a dual-loop TES configuration. Thermodynamic design and performance of the system are modelled under realistic assumptions for all components. A parametric analysis is then conducted to investigate the effects of key parameters on the overall system performance. Results show that the system can achieve a round-trip efficiency of 74.7% and a storage density of 61.1 kWh/m³, provided that an appropriate seawater depth is available. This study offers valuable insights for advancing CCES systems.