Seasonal energy storage is an important component to cope with the challenges resulting from fluctuating renewable energy sources and the corresponding mismatch of energy demand and supply. The storage of heat via medium deep borehole heat exchangers is a new approach in the field of Borehole Thermal Energy Storage. In contrast to conventional borehole storages, fewer, but deeper borehole heat exchangers tap into the subsurface, which serves as the storage medium. As a result, the thermal impact on shallow aquifers is strongly reduced mitigating negative effects on the drinking water quality. Furthermore, less surface area is required. However, there are no operational experiences, as the concept has not been put into practice so far. In this study, more than 250 different numerical storage models are compared. The influence of the characteristic design parameters on the storage system's behaviour and performance is analysed by variation of parameters like borefield layout, fluid inlet temperatures and properties of the reservoir rocks. The results indicate that especially larger systems have a high potential for efficient seasonal heat storage. Several GWh of thermal energy can be stored during summertime and extracted during the heating period with a high recovery rate of up to 83%. Medium deep borehole heat exchanger arrays are suitable thermal storages for fluctuating renewable energy sources and waste heat from industrial processes. Copyright © 2016 John Wiley & Sons, Ltd.

Heating of buildings requires more than 25% of the total end energy consumption in Germany. By storing excess heat from solar panels or thermal power stations of more than 110°C in summer, a medium deep borehole thermal energy storage (MD-BTES) can be operated on temperature levels above 45°C. Storage depths of 500 m to 1,500 m below surface avoid conflicts with groundwater use. Groundwater flow is decreasing with depth, making conduction the dominant heat transport process. Feasibility and design criteria of a coupled geothermal-solarthermal case study in crystalline bedrock for an office building are presented and discussed

Performance and accuracy of machine learning techniques to segment rock grains, matrix and pore voxels from a 3-D volume of X-ray tomographic (XCT) grayscale rock images was evaluated. The segmentation and classification capability of unsupervised (k-means, fuzzy c-means, self-organized maps), supervised (artificial neural networks, least-squares support vector machines) and ensemble classifiers (bragging and boosting) were tested using XCT images of andesite volcanic rock, Berea sandstone, Rotliegend sandstone and a synthetic sample. The averaged porosity obtained for andesite (15.8 ± 2.5 %), Berea sandstone (16.3 ± 2.6 %), Rotliegend sandstone (13.4 ± 7.4 %) and the synthetic sample (48.3 ± 13.3 %) is in very good agreement with the respective laboratory measurement data and varies by a factor of 0.2. The k-means algorithm is the fastest of all machine learning algorithms, whereas a least-squares support vector machine is the most computationally expensive. Metrics entropy, purity, mean square root error, receiver operational characteristic curve and 10 K-fold cross-validation were used to determine the accuracy of unsupervised, supervised and ensemble classifier techniques. In general, the accuracy was found to be largely affected by the feature vector selection scheme. As it is always a trade-off between performance and accuracy, it is difficult to isolate one particular machine learning algorithm which is best suited for the complex phase segmentation problem. Therefore, our investigation provides parameters that can help in selecting the appropriate machine learning techniques for phase segmentation.