Cèsar Fernández

This paper presents a practical case study, conducted with Corporación Alimentaria Guissona S.A. (bonÀrea), on optimizing storage locations in a large-scale, automated warehouse for perishable foods. The objective is to minimize costly blockages on the facility’s conveyors, which disrupt throughput and risk the cold chain. We address this via item redistribution. This is a hard combinatorial optimization problem with a large search space. We model the problem using a derivative-free simulator and explore solutions using local search metaheuristic techniques. Unlike conventional optimization that seeks global optima in an unrestricted search space, our approach is limited by a defined local distance from the original solution. We show different approaches to tackle this set of problem constraints based around local search metaheuristics that ensure bounded exploration of the solution space. Our results demonstrate that a bounded exploration approach can reduce blockages by 45%, providing a practical, low-disruption solution that significantly improves operational efficiency in a real-world, high-volume logistics environment.

The generation of synthetic data is a critical area of research in domains where real data are either not available in large quantities or cannot be directly used. Different techniques have been developed to produce high-quality, realistic synthetic datasets which retain the statistical properties of the original data. State-of-the-art results focus on the use of neural networks to capture the latent space extracted from the data. While recent advances in the field of deep learning motivate the application in this new context, this paper proposes a naive baseline to generate synthetic data based on pairwise probabilities. We name the technique DISCO (Discrete Intersection Synthesizer through Combinatorial Optimization), a novel synthetic generator for tabular data. DISCO models data by optimizing the intersection of pairwise probabilities on each generated row in order to resemble the original dataset. Our approach preserves marginal (and pairwise) distributions and as a result, resembles the original data with high fidelity with a very simple approach. Evaluation on various synthetic and real-world datasets as well as regression and classification tasks prove DISCO’s ability to generate high-quality data that rivals state-of-the-art models in both statistical accuracy and machine learning efficacy.

Solar thermal energy coupled to a seasonal sorption storage system stands as an alternative to fossil fuels to supply residential thermal energy demand in climates where solar energy availability is high in summer and low in winter, matching with a high space heating demand. Sorption storage systems usually have a high dependency on weather conditions (ambient temperature and solar irradiation). Therefore, in this study, the technical performance of a solar-driven seasonal sorption storage system, using an innovative composite sorbent and water as working fluid, was studied under three European climates, represented by: Paris, Munich, and Stockholm. All scenarios analyses were simulation-based under optimal system control, which allowed to maximize the system competitiveness by minimizing the system operational costs. The optimal scenarios profit from just 91, 82 and 76% of the total sorption system capacity, for Paris, Munich, and Stockholm, respectively. That means that an optimal control can identify the optimal sorption storage size for each location and avoid oversizing in future systems, which furthermore involves higher investment costs. The best coefficient of performance was obtained for Stockholm (0.31), despite having the coldest climate. The sorption system was able to work at minimum temperatures of −15 °C, showing independence from ambient temperature during its discharge. In conclusion, a seasonal sorption system based on selective water materials is suitable to be integrated into a single-family house in climates of central and northern Europe as long as an optimal control based on weather conditions, thermal demand, and system state is considered.

Deep Reinforcement Learning (DRL) proved to be successful for solving complex control problems and has become a hot topic in the field of energy systems control, but for the particular case of thermal energy storage (TES) systems, only a few studies have been reported, all of them with a complexity degree of the TES system far below the one of this study. In this paper, we step forward through a DRL architecture able to deal with the complexity of an innovative hybrid energy storage system, devising appropriate high-level control operations (or policies) over its subsystems that result optimal from an energy or monetary point of view. The results show that a DRL policy in the system control can reduce the system operating costs by more than 50%, as compared to a rule-based control (RBC) policy, for cooling supply to a reference residential building in Mediterranean climate during a period of 18 days. Moreover, a robustness analysis was carried out, which showed that, even for large errors in the parameters of the system simulation models corresponding to an error multiplying factors up to 2, the average cost obtained with the original model deviates from the optimum value by less than 3%, demonstrating the robustness of the solution over a wide range of model errors.

In this study, a thermal energy storage tank filled with commercial phase change material flat slabs is investigated. The tank provides heat at around 15 °C to the evaporator of a seasonal thermal energy storage system developed under the EU-funded project SWS-Heating. A 2D numerical model of the phase changed material storage tank based on the finite control volume approach was developed and validated with experimental data. Based on the validated model, an optimization was performed to identify the number, type and configuration of slabs. The final goal of the phase change material tank model is to be implemented into the whole generic heating system model. A trade-off between results’ accuracy and computational time of the phase change material model is needed. Therefore, a comparison between a 2D implicit and 2D explicit scheme of the model was performed. The results showed that using an explicit scheme instead of an implicit scheme with a reasonable number of nodes (15 to 25) in the heat transfer fluid direction allowed a considerable decrease in the computational time (7 times for the best case) with only a slight reduction in the accuracy in terms on mean average percentage error (0.44%).