Sammendrag
With its long coastline and deep fjords, Norway has over 130 ferry connections. Both governmental restrictions and international goals propel the ferry industry to supply these ferry connections with low- and zero-emission ferries. To compete with cheap, lightweight diesel machinery, zero-emission vessels need to be custom-made not to sacrifice performance to reduce emissions. It is challenging to obtain all the relevant information to optimize the zero-emission vessel design. Therefore, finding methods for generating this information is the key to success. This thesis aims to find methods for generating relevant information related to the optimal design problem. This problem includes generating route information from AIS data, finding an energy model that can be parameterized by available information, and providing technical information such as equipment efficiency, weight, and volume. All this, in combination, provides the background for finding a custom-made design of a zero-emission fast ferry. In this thesis, two sets of AIS data have been studied. The first one, from 2018, was characterized by inconsistent sampling time and wild-points in position and speed. The second set of AIS data, from 2021, had significantly higher quality with 1000 times higher resolution. From this data, an implementation of a DBSCAN clustering algorithm was able to detect ports, which further led to the generation of the tracks of the route. Furthermore, the average speed profile from the route processing provided a good foundation for energy modeling along the route. This provided valuable information to the optimal design problem. In this thesis, the zero-emission technology of interest for short-distance operations of fast ferries are batteries and fuel cells. The gravimetric density of batteries is much higher than for hydrogen, but due to the weight of the fuel cell stack, the choice of equipment should be made with care. This thesis conducts a case study of MS Tyrhaug, operating the ferry connection between Trondheim and Kristiansund. AIS data for this ferry was investigated, and the results show that AIS data is mature for providing route information. Evaluation of efficiency, weight, and volume of a battery configuration and a fuel cell configuration shows that a fuel cell configuration is the most suitable for longer distances, without charging opportunities, and battery configuration for shorter distances. The limit is found to be around 200 nm. A battery configuration for the case vessel, MS Tyrhaug, will be the most appropriate if the battery can be fully charged at both end ports. The battery's capacity should be at least 2660 kWh, with a weight and volume of 19,000 kg and 17.6 $m^3$, respectively. A hybrid-electric configuration has also been investigated, consisting of a fuel cell and a battery. An implementation of a dynamic programming (DP) algorithm has been applied to find a global solution to the optimal loadsharing problem. The DP implementation is simulated with different cost functions emphasizing various parameters to enlighten the importance of tuning. For example, when emphasizing 100\% with the fuel consumption, the fuel consumption was reduced by 10 kg compared to a simulation emphasizing all parameters of interest. This illustrates the importance of a well-functioning loadsharing system.
Vis fullstendig beskrivelse
