Pilot 2: Globally Optimal Design of Ship Energy Systems
Ship energy system conceptual design happens at very early phase of ship design when there is only limited amount of information available. Consequently, large uncertainties exist due to the unknown information and inaccurate estimations. This is usually addressed by design margins, which causes over-dimensioning of or imbalance in subsystems and therefore suboptimal design concepts of ship energy systems. Even worse, due to the limited consideration of real operational environment during the design phases, the already suboptimal ship energy systems are often operated off the design points, which consequently leads to lower operational efficiency, and higher life cycle costs and emissions. On the other hand, both model-based and data-driven approaches have been only playing supporting roles in the design of ship energy systems so far. The full potential of simulation methods and huge amount of design and operational data still remains untapped.
Use Cases
During the first evaluation period of the project, the focus of the pilot was on the following use cases
Ship Energy Systems Design through Digital Twins
VTT incorporates AI, Big Data analytics, HPC and hybrid digital twin in the loop of ship energy system design and optimization process so as to achieve more energy- and cost-efficient conceptual designs in both design and practical operations throughout lifecycles, with reduced time, efforts and costs.
Ship Energy Systems Optimisation
VTT’s work on novel maritime technologies addresses the challenges of reducing emissions and increasing the efficiency of global shipping, according to the international targets set by IMO.
Pilot-participating Partners
Pilot Technological Highlights
The technological highlights of this pilot were the following
Ship design diagrams, onboard sensor data, AIS data, noon reports, key subsystem project guides.
Data
Ship energy system conceptual design problems are formulated as the determination of a set of design variables (e.g. deadweight, installed power at main
engines and auxiliary engines, shaft generator, exhaust boilers, propulsion, fuel input, heat and power demand) that results in a number of design scenarios
satisfying certain pre-defined targets, conditions and requirements. It is a class of combinatorial problem which can be addressed as the selection and
configuration of a set of design variables satisfying multiple criteria specified (e.g. energy efficiency, cost, emissions).
Components
The complexity of the multi-criteria problem increases exponentially with the growth of the number of the design variables and the nonlinearity of the
dynamics and uncertainty of operating conditions. Therefore, thousands and even millions of design scenarios within full design space need to be evaluated
and a huge amount of data need to be utilised so as to find the truly global optimal among the design alternatives. Given that there are usually more than
1 million of alternatives for one full-scale case study, an efficient framework is needed to effectively tackle the compute-intensive, and highly complex
and dynamic, energy system design problems to find the optimal solutions in less time.
Highlights
Pilot Results
VTT is formulating a beyond-the-state-of-the-art framework into a web-based digital service platform for ship energy system conceptual design and optimization, with extended capabilities in AI, big data analytics and HPC. It can generate and handle a large number of design scenarios in hours, instead of days and weeks, leaving a ground-breaking potential for the exploration and experimentation in the conceptual designs. This will help not only to efficiently find globally optimal design and operation of ship energy systems but also to significantly minimize the time, efforts and costs of ship energy system conceptual design and optimization process.