Quays Rather Than Boilers

Abstract

As part of the effort to make the Dutch waterways CO2-neutral, a consortium consisting of, among others, CRUX Engineering BV and Delft University of Technology, initiated a case study in which the thermal and mechanical performance of an energy sheet pile wall is studied. This technology allows for the extraction of energy from the water and soil by means of collectors, i.e. a system of pipes welded to the sheet piles. The fluid flowing through these pipes is directed to a heat pump after which the extracted energy can be used to heat buildings. Moreover, during summer, the system provides the opportunity to be used to cool homes by storing heat in the subsurface. The primary objective of a sheet pile wall is to assure stability of the quay and by giving the sheet pile wall an additional purpose, that of heat sink or source, previously not present thermal loads are applied to the wall. In this thesis the temperature change of the sheet pile wall and soil is investigated after which the deformation of the sheet pile wall is analysed in order to answer the main research question: 'Can the structural safety of a sheet pile quay be guaranteed when the sheet pile is thermally activated?'.

In the town of Zweth, part of the municipality of Rotterdam, an investigation into the performance and consequences of thermally activating a sheet pile wall was conducted in the form of a case study. A sheet pile wall with a width of approximately 7.5 meters and a depth of 15 meters was placed. Two sets of loops were welded on the sheet pile: loops running to the bottom of the wall and loops running to a depth of 3 meters. The influence on the surroundings of these separately investigated loops was monitored by means of thermistor strings. The deformation of the sheet pile was measured with inclinometers. This thesis starts with the analysis of the collected data from the field test. To investigate the long-term behaviour of the sheet pile wall two models are developed in this thesis: a heat conduction model and a geotechnical stability model. The former is a thermal calculation only and aims to model the temperature changes in the subsurface as well as the heat extraction. The latter consists of a time dependent Thermo-Hydro-Mechanical (THM) analysis in PLAXIS and used the sheet pile temperature determined in the heat conduction model as thermal boundary condition on the sheet pile. The resulting temperature change of the sheet pile and subsurface in combination with mechanical loads results in a deformation of the sheet pile. Using the data acquired in the field test, these two models are validated. With the help of the validated geotechnical stability model, the thermal component in the total distortion of the sheet pile is investigated. Finally, a calculation with with a time span of 4 years is performed to see the long-term reaction of the sheet pile wall. Based on this research the main question is answered.

In the Netherlands, soil temperatures below 6 meters of depth stays approximately 12 degrees Celsius all year. Above that, the soil temperature changes with the seasons, yet with a delay. The test ran in the autumn and winter and due to the time of the year, the soil at a few meters depth still had the summer heat stored while the soil close to the surface already started to cool down. The interpretation of the collected data started with recognizing from the thermistor string data that the soil in the quay adjacent to the canal is heavily influenced by the water temperature. Deeper down, the heat extraction is the only external influence. Investigations into heat extraction based on the numbers of loops activated showed that there was no linear relation between the concentration of activated loops and the soil temperature decrease. When twice as many loops were activated on the same sheet pile wall area, the soil temperature decrease was only slightly higher. The maximum deformation of the sheet pile wall oscillated throughout the testing period. No direct correlation between thermal activation and deformation was identified.

The validation of the heat conduction model showed that the the soil temperatures in the model were generally too low in the subsurface below 6 meters in the comparison between the in-situ measured temperatures and the numerically determined soil temperatures. The numerically approached temperatures in the quay adjacent to the canal were strongly influenced by the canal water temperature. So much that the heat extraction was barely visible in the data. This emphasizes the importance of good water temperature data. Moreover, part of the energy extracted is a result of the turbulent flow of the canal water parallel to the sheet pile wall. This is not simulated in the model yet important to add. Despite this, the numerically determined extracted heat and the measured heat showed a reasonably good fit.

The comparison of in-situ measured deformation of the sheet pile wall and the numerically approached deformation was in the same order of magnitude as the measured data. The influence of the thermal component on the distortion of the sheet pile was found to be small. During the calculation, the distortion slightly oscillated in the range of tens of millimetres which falls within the numerical accuracy of the model. The measured oscillation was assumed to be caused by changing water levels and soil saturation as well as measurement errors. To investigate the long-term reaction of a thermally activated sheet pile wall, three THM computations with a time span of 4 years were conducted: one without thermal activation, one with natural soil temperature regeneration during summer and one with forced regeneration by the energy sheet pile wall. All had a sinus-like course with an accumulating deformation over time which was intensified by the amount of influence of the energy sheet pile. The computation with the forced regeneration -- i.e. the most influence -- by the energy sheet pile wall ended with the largest amount of deformation which was 5 millimetres larger than the computation without thermal activation. Extrapolating the results, the maximum deformation was determined to be approximately 6 millimetres after 6 years.

The thermal load on the sheet pile wall is not a design parameter and is generally not investigated in detail. Therefore, it is strongly advised to conduct additional research. However, based on these findings, it is concluded that the deformation resulting from the thermal activation of a sheet pile wall is not excessively large. This means the primary objective of keeping the quay stable won't be compromised. This leads to the conclusion that the technology is safe to be applied on a large scale to decrease the carbon footprint and contribute to a more sustainable society without interfering with the quay's primary function.