• Energy Research
• other engineering and technologies close

This study evaluates the potential of supervised and transfer learning techniques to assist energy system optimization. A surrogate model is developed with the support of a supervised learning technique (by using artificial neural network) in order to bypass computationally intensive Actual Engineering Model (AEM). Eight different neural network architectures are considered in the process of developing the surrogate model. Subsequently, a hybrid optimization algorithm (HOA) is developed combining Surrogate and AEM in order to speed up the optimization process while maintaining the accuracy. Pareto optimization is conducted considering Net Present Value and Grid Integration level as the objective functions. Transfer learning is used to adapt the surrogate model (trained using supervised learning technique) for different scenarios where solar energy potential, wind speed and energy demand are notably different. Results reveal that the surrogate model can reach to Pareto solutions with a higher accuracy when grid interactions are above 10% (with reasonable differences in the decision space variables). HOA can reach to Pareto solutions (similar to the solutions obtained using AEM) around 17 times faster than AEM. The Surrogate Models developed using Transfer Learning (SMTL) shows a similar capability. SMTL combined with the optimization algorithm can predict Pareto fronts efficiently even when there are significant changes in the initial conditions. Therefore, STML can be used along with the HOA, which reduces the computational time required for energy system optimization by 84%. Such a significant reduction in computational time enables the approach to be used for energy system optimization at regional or national scale.

Higher availability of future climate data sets, generated by regional climate models (RCMs) with fine temporal and spatial resolutions, improves and facilitates the impact assessment of climate change. Due to significant uncertainties in climate modeling, several climate scenarios should be considered in the impact assessment. This increases the number of simulations and size of data sets, complicating the assessment and decision making. This article suggests an easy-to-use method to decrease the number of simulations for the impact assessment of climate change in energy and building studies. The method is based on synthesizing three sets of weather data out of one or more RCMs: one typical and two extremes. The method aims at decreasing the number of weather data sets without losing the quality and details of the original future climate scenarios. The application of the method is assessed for an office building in Geneva and the residential building stock in Stockholm.Results show that using the synthesized data sets provides an accurate estimation of future conditions. Variations and uncertainties of future climate are represented by the synthesized data. In the case of synthesizing weather data using multiple climate scenarios, the number of simulations and the size of data sets are decreased enormously. Combining the typical and extreme data sets enables to have better probability distributions of future conditions, very similar to the original RCM data.

In recent years, climate change and the corresponding expected extreme weather conditions have been widely recognized as potential problems. The building industry is taking various actions to achieve sustainable development, implement energy conservation strategies, and provide climate change mitigation. In addition to mitigation, it is crucial to adapt to climate change, and to investigate the possible risks and limitations of mitigation strategies. Although the importance of climate change adaptation is well-understood, there are still challenges in understanding and modeling the impacts of climate change, and the consequent risks and extremes. This work provides a comprehensive study of the impacts of climate change on the energy performances and thermal comfort of European residential building stocks. To perform an unbiased assessment and account for climate uncertainties and extreme events, a large set of future climate data was used for a 90-year period (2010–2099). Climate data for 38 European cities in five different climate zones, downscaled by the “RCA4” regional climate model, were synthesized and applied to simulate the respective energy performances of the residential building stocks in the cities. The results suggest that there will be larger needs for cooling buildings in the future and less heating demand; however, there are differences in the variation rates between zones and cities. Discomfort hours will increase notably in cities within cooling-dominated zones, but will not be affected considerably in cities within heating-dominated zones. In addition to long-term changes, climate-induced extremes can considerably affect future energy demands, especially the cooling demand; this may become challenging for both buildings and energy systems.

AbstractSince the beginning of the 21st century, straw-bale buildings are reappearing in the world; however, their thermal performances were not thoroughly investigated up to now. The purpose of this study is to analyze thermal behavior and energy performance of a straw-bale building in Switzerland. Using Pleiades+Comfie Software, building designs have been studied to understand the best way to mitigate overheating risks due to the low heat capacity of straw. Thermal-dynamic results and Life Cycle Assessment conclude that straw bale buildings can be a sustainable alternative in the energy evolution of building construction, due to its low embodied energy and excellent thermal performance.

Distributed energy systems play a significant role in the integration of renewable energy technologies. The Energy Internet links a fleet of distributed energy systems to each other and with the grid. Interactions between the distributed energy systems via information sharing could significantly enhance the efficiency of their real-time operation. However, privacy and security concerns hinder such interactions. A game-theoretic approach can help in this regard, and enable consideration of some of these factors when maintaining interactions between energy systems. Although a game-theoretic approach is used to understand energy systems' operation, such complex interactions between the energy systems are not considered at the early design phase, leading to many practical problems, and often leading to suboptimal designs. The present study introduces a game-theoretic approach that enables consideration of complex interactions among energy systems at the early design phase. Three different architectures are considered in the study, i.e., energy eystem prior to grid (ESPG), fully cooperative (FCS), and non-cooperative (NCS) scenarios, in which each distributed energy system is taken as an agent. A novel distributed optimization algorithm is developed for both FCS and NCS. The study reveals that FCS and NCS reduce the cost, respectively, by 30% and 15% compared to ESPG. In addition to cost reduction, there is a significant change in the energy system design when moving from FCS to NCS scenarios, clearly indicating the requirement for a scenario that lies between NCS and FCS. This will lead to reducing design costs while maintaining privacy.

An integrated approach is presented in this study to design electrical hubs combining optimization, multi-criterion assessment and decision making. Levelized Energy Cost (LEC), Initial Capital Cost (ICC), Grid Integration Level (GI), Levelized CO2 emission (LCO2), utilization of renewable energy, flexibility of the system, loss of load probability (LOLP) are considered as criteria used to assess the design. The novel approach consists of several steps. Pareto analysis is conducted initially using 2D Pareto fronts to reduce the dimensions of the optimization problem. Subsequently, Pareto multi objective optimization is conducted considering LEC, GI and ICC which were identified as the best set of objective functions to represent the design requirements. Next, fuzzy TOPSIS and level diagrams are used for multi-criterion decision making (MCDM) considering the set of criteria and the boundary matrix that represents the design requirements of the application. Pareto analysis shows that 5D optimization problem can be reduced to a 3D optimization problem when considering LEC, ICC and GI as the objective functions. Finally, results obtained from the case study shows that the novel method can be used design distributed energy systems considering a set of criteria which is beyond the reach of Pareto optimization with different priority levels.

Extreme climate events occur more frequently and stronger in the future due to climate change. Maintaining the energy security during extreme conditions is essential to reduce the impacts of extreme climate and avoid disasters. Resilient design of the energy system to resist against extreme climate events are investigated considering four scenarios, namely, typical demand (TD), extreme demand (ED), extreme renewable energy generation (ER) and, extreme demand and renewable generation (EDR). A regional climate model is used to develop the four scenarios with the assistance of a building simulation model. Subsequently, multi-energy hub is optimized for each scenario considering net present value (NPV) and grid integration (GI) level as the objective functions. A significant difference in objective function values is observed when analyzing the four scenarios. Similarly, a significant difference in the energy system design is observed when moving from one scenario to another. The results of the study reveal that a energy system design is strongly influenced by extreme climate scenario considered which will make the energy system to be a sub-optimal when operating at a different climatic condition with a significant performance gap. Therefore, improving the climate flexibility of energy systems is an essential task which is challenging at the early design process.

Neglecting extremes and designing buildings for the past or most likely weather conditions is not the best approach for the future. Robust design techniques can, however, be a viable option for tackling future challenges. The concept of robust design was first introduced by Taguchi in the 1940s. The result of the design process is a product that is insensitive to the effect of given sources of variability, even though the sources themselves are not eliminated. A robust design optimization (RDO) method is for the first time proposed in this paper, for supporting architects and engineers in the design of buildings with robust energy performance under climate change and extreme conditions. The simplicity and the low computational demand of the process underlies the feasibility and applicability of this method, which can be used at any stage of the design process. The results show that the performance of the optimum solution not only has a 81.5% lower variation (less sensitivity to climate uncertainty) but at the same time has a 14.4% lower mean energy use value compared with a solution that is compliant with a recent construction standard (ASHRAE 90.1-2016). Less sensitivity to climate uncertainty means greater robustness to climate change whilst maintaining high performance.