Improved Estimations of Energy Consumption for Dredging Activities Based on Actual Data

More Info


The consequences of climate change are happening now, and will increase in the years to come. Understanding the energy consumption of dredging operations is becoming increasingly important for dredging contractors such as Van Oord. With the emergence of more complex dredging situations, including dredging at limited depths and short sailing distances, it has become more difficult to provide an accurate estimate. This research was conducted to evaluate Van Oord's current estimation method, called VOHOP, and to provide an improved estimate of the energy consumption for dredging activities based on historical data. Therefore, the main objective of the study is as follows:

"To improve the estimation of energy consumption for dredging activities based on actual data."

To achieve this goal, the method that is selected for creating improvement is comparing estimation methods with actual data where the difference gives insight in how to improve the estimation method, explained more briefly: EST vs ACT = DIFF. A literature review is first conducted to indicate what a dredging cycle looks like. After analyzing the different phases, consisting of sailing empty, loading, sailing full and disposing, the parameters that play a role in the consumption of energy during the dredging process are stated. For example for the sailing stages parameters that play a role are: sailing distance and sailing speed. Due to the relatively less elaborate calculation of energy consumption for the sailing stages, an improvement of estimation accuracy for these phases is considered to be valuable. Next, existing alternative methods for estimating energy consumption for the sailing phases are explored and the method proposed by Holtrop & Mennen (HM), including modifications created for limited depth estimations, is selected as the most appropriate because it takes into account a larger number of parameters (such as depth, draught and distance), which is expected to increase the accuracy. After comparing the current estimation method VOHOP, with the method proposed by HM including modifications, one of the main findings is that the modified HM approach has a lower outcome of energy consumption compared to VOHOP especially for larger depths. After comparing, the starting point of the selected estimation method (EST) is created. The comparison of the modified HM method and VOHOP is done by means of the simulation software OpenCLSim in Python.

To validate the accuracy of either VOHOP or modified HM, comparative material consisting of actual data provided by Van Oord is used. Multiple projects (20+) that have been performed by a trailing suction hopper dredger (TSHD) of Van Oord are analyzed after which three are selected. The selection is based on the amount of data available for a project and the difference in depth and distance, to use data that is suitable for comparison. After filtering out non-representative data (missing data points or too large time steps), the actual data is analyzed where a distinction is made between acceleration, free sailing and deceleration. Using the suitable data for comparison contributes to the actual data (ACT) part of this research.
The selected actual data is compared with the modified HM approach for estimating energy consumption, which contributes to the DIFF part. For the free sailing part, a main finding when comparing the modified HM with the actual data is the deviation (lower estimation compared to actual data) for increasing depth estimations (analyzed for depths up to 60 m). A possible reason for this is due to the applied modifications for the modified HM approach that are observed to have less impact on the estimation for increasing depths with respect to the draught. However, estimates for limited depth also show a deviation as the results are more sensitive and show an asymptotic behaviour for which a correction is recommended. It is therefore concluded that the modified HM approach needs corrections for both estimating deeper situations as for situations with more limited depths. After analyzing the resistance components of the modified HM approach in different situations, consisting of different ratios of depth and draught, the reason for the large deviation in the modified HM method for deeper situations and the greater sensitivity for more limited depths is found in the contribution of the wave resistance term $R_w$. This is found out after comparing and determining the contribution of all resistance terms for multiple situations gathered from real time projects of actual data containing increasing depths and draughts. A correction to the wave resistance calculation is made using actual data, finding a fitting between increasing wave resistance for increasing vessel draught with respect to water depth. The result is a proposed HM method containing a less sensitive estimation method that takes into account the draught of the vessel related to the water depth for both deeper situations and shallower situations.

The acceleration and deceleration phases of the actual data are considered inappropriate for comparison with the modified HM approach, since the used power in the actual data contains the effect of slip, causing the vessel to use more power for reaching a certain velocity than actually required according to modified HM. Therefore, to improve the estimation of the acceleration and deceleration, a generalized fit is found in the actual data for the power in relation to time. This fit is generated for different ranges of depth and is coupled with the sailed distance to quantify whether the distance has any effect on the results. It is found that at a short sailing distance a significantly lower velocity is achieved, which in turn affects the power used. By taking into account depth and distance, an estimation fit results that takes into account a larger number of parameters (velocity, distance, depth and draught) than the VOHOP method, and is therefore expected to produce more accurate results. With the estimation fits created for acceleration and deceleration and the corrected estimation method, HM proposed, for free sailing, the outcome is a newly created estimation tool that is based on historical data.

The improvement of the newly developed estimation method is validated by comparing VOHOP and the new estimation tool with eight cases from historical data. The cases are selected at different depths and different distances and are found in the selected projects for this research. The results show an increase in accuracy for the acceleration phase, while the other two phases, free sailing and deceleration, show fluctuating results. The reason for the fluctuation in the free sailing phase is the difference in duration of the phase caused by the definition of acceleration and free sailing, where the new estimation tool considers part of the free sailing as acceleration. The results indicate that the power usage is more accurately estimated by the new estimation tool, which indicates that the fluctuations of energy estimates are caused by a miss-estimation of duration instead of power usage. For deceleration, the fluctuating results can be attributed to the fluctuating pattern of deceleration, which may be caused by external factors such as traffic or currents, making it difficult to predict the energy consumption for this part of the sailing phase. All in all, the new estimation tool exhibits greater accuracy for the overall energy consumption estimation compared to the current estimation method VOHOP. In addition, the estimation of power consumption during free sailing shows improvement, indicating that incorrect estimates of energy during free sailing are more likely to be caused by inaccurate estimations of the duration instead of inaccurate estimations of power.

In conclusion, the results show that when considering depth, distance and draught as input parameters and applying a data based correction, the accuracy of the estimation method increases. Where the results do show an improvement, it is recommended that further research be conducted on the estimation of energy consumption and sailing duration for acceleration and deceleration. Several assumptions were made in this study to simplify the analysis such as constant wave climate, where a follow-up study could verify these assumptions or even improve the results of the study by implementing additional consequences for these assumptions. Moreover, this study took into account a certain number of parameters and observed only one vessel type, where adding more parameters or considering more vessel types would give a clear picture of the reliability of the estimation method and energy consumption. A further study is needed to perform the same procedure of improving the estimation also for the other phases of a dredging cycle, since these phases do contribute to a large extent of the total energy consumption of a dredging cycle. Finally, using a larger amount of data for this study could reduce the uncertainty of the results and increase the reliability. In addition, an increase in the quality of the data further increases the reliability of the results.