TECHNICAL FIELD

The present disclosure relates to methods and control units for determining a marine vessel hybrid driveline control strategy, i.e. , to determine target power outtake levels along a route from two or more driveline types in a marine hybrid driveline arrangement.

BACKGROUND

Emission reduction, specifically emission of carbon dioxide (CO2), is an important focus area in many transportation industries, including marine cargo coastal transport and ferries.

Hybrid driveline technology is a known method for reducing unwanted emission. In this case a conventional driveline type, such as a diesel-based combustion engine driveline, is complemented by a second driveline type such as an electric driveline which has a reduced emission level compared to the first driveline type. This way the peak performance of the vessel is left intact, since the conventional driveline type is there, while allowing for reduced emission operation by the second driveline type. The two driveline types of the vessel propulsion system normally differ in the amount of available energy. A battery system for instance often holds less energy when fully recharged compared to a full fuel tank of a combustion engine-based driveline.

US 2011/0320073 A1 relates to energy optimization in marine vessels.

Despite the progress made to-date, there is still a need for improvement in marine energy optimization.

SUMMARY

It is an object of the present disclosure to provide improved methods for controlling marine vessel hybrid drivelines in a marine vessel propulsion system. This object is obtained by a computer implemented method for determining a marine vessel hybrid driveline control strategy. The hybrid driveline of the vessel propulsion system comprises at least a first driveline type and a second driveline type different from the first driveline type. One of the driveline types often has a more limited amount of available energy compared to the second driveline type, as mentioned above. For instance, an electrical driveline type is limited by the energy in the electrical energy storage system, which normally holds much less energy compared to that held in a Diesel fuel tank or the like. The control strategy comprises a power outtake from the first driveline type and from the second driveline type along a route to be traversed by the marine vessel. The control strategy also has an associated emission target. The method comprises determining a total power requirement for the marine vessel along the route to be traversed, obtaining a power-to-emission relationship for each of the driveline types, where a power-to-emission relationship is indicative of an emission by the driveline type when operating at a given power outtake, and determining the marine vessel hybrid driveline control strategy based on the total power requirement for the route and on the power-to-emission relationships, such that the sum of power outtakes from the driveline types is greater or equal to the total power requirement along the route and such that the sum of emissions from the driveline types meets the emission target. Thus, a balance between power outtakes from the two or more driveline types is determined which meets a set of control targets, among which is emission. The emission target can be a pre-determined emission amount for the route and/or a minimization of emission along at least a part of the route. A driveline type may comprise any of a diesel-based driveline, a gasoline-based driveline, a fuel cell electric system, a hydrogen-based combustion driveline, a battery-powered electric driveline, and a liquefied natural gas (LNG) based driveline. The emission by a driveline type may comprise, e.g., emission of carbon dioxide, CO2, carbon monoxide, CO, nitrogen oxide, NOx, and/or noise.

The total available energy for one or both driveline types over the route to be traversed by the vessel may be limited and distance (or time) to the next available energy replenishment opportunity together with estimated time of energy replenishment can be considered in the methods discussed herein.

The route to be traversed by the marine vessel is normally an a-priori known route, i.e. , a deterministic path which the vessel is expected to follow with high probability. This is for instance the case if the vessel is a ferry which travels along a predetermined route in a repetitive manner. This is also the case if the vessel is a coastal transport vessel with a fixed route or set of routes that it follows over and over again, or at least from time to time. Some of the methods disclosed herein leverage such repetitive travel patterns in order to improve the control strategy. In a way, some of the methods disclosed herein can be seen as methods for iterative improvement of a hybrid driveline control strategy, where the method, or some model used in the method, is improved every time the vessel travels along the route.

Routes determined by route planning methods based on a known start location and destination, as well as routes determined at least partly based on a current location of the marine vessel, on a prediction of destination based on previous routes traversed by the marine vessel, and on a route planning method based on current location and predicted destination, are also supported by at least some of the methods disclosed herein. The route prediction can be made even more accurate if the time of day and/or date is considered in the route inference. A route travelled up to a given location can also present valuable input data to the route inference, since vessels often return via the same route that was taken to the current location. This is especially true for day-cruisers and other smaller marine leisure craft. According to some aspects, the method also comprises updating the determined marine vessel hybrid driveline control strategy in case the destination of the route changes. The a-priori known route is often known not only in terms of its path on a map, but together with the width of the route (“highway”) at any given point.

The route to be traversed by the marine vessel is preferably associated with one or more environment conditions of one or more route segments. The environment conditions may comprise, e.g., wave height, sea current, and wind conditions along the route. The forecasts can be calibrated, i.e., corrected, with data from local sensors on the vessel, such as wind sensors and wave height sensor systems based on cameras and/or based on an inertial measurement unit (IMU). This calibration can also be performed in an iterative manner, i.e., updated every time the vessel travels along a given route. For example, consider a vessel such as a ferry which repetitively travels along a route. The vessel may compare a wind forecast provided for a larger area to actual wind conditions along the route, every time the vessel traverses the route. The vessel is then in a position to identify locations along the route where the local wind conditions deviate from the wind forecast for the larger area, and thus construct a calibration table which can be applied to improve accuracy of the wind conditions along the particular route.

According to some aspects, the method comprises determining the total power requirement at least in part based on a recommended speed profile of the marine vessel along the route. The recommended speed profile can, for instance, be set based on target travel time or based on a target arrival time. Generally, the higher the speed the higher the power requirement to complete the route. Hence, the use of a target speed profile along the route as input to the control system could increase performance of the control strategy.

The method may also comprise determining the total power requirement at least in part based on a target noise level of the hybrid driveline along the route. This allows for reduction of noise level, e.g., along parts of a route associated with requirements on noise emission, which is an advantage.

According to some aspects, the method comprises determining the power-to- emission relationship at least partly based on an obtained efficiency for each driveline type and on respective constant conversion factors for each driveline type. This way the optimization procedure can be tailored for a given set of driveline types to increase accuracy, which is an advantage. The method may also comprise determining the marine vessel hybrid driveline control strategy as a solution to an optimization problem with an objective function formulated in terms of emission for a given power outtake distribution along the route. This solution can be obtained in several different ways, e.g., as the solution to the optimization problem from a machine learning (ML) algorithm, and/or from evaluating a pre-determined set of candidate control strategies. These methods have been shown to be particularly efficient when applied to vessel that travel along routes in a repetitive manner, such as ferries or coastal transport vessels. This is because the same route is travelled over and over again, allowing an ML algorithm to be trained for high accuracy predictions. A relatively large number of candidate control strategies can also be evaluated since the same trip is repeated over and over. Thus, the repetitive nature of travel for a given vessel can be exploited in order to iteratively refine the control strategy for a given trip.

The method may also comprise determining the marine vessel hybrid driveline control strategy based on one or more constraints, where the constraints comprise any of: a maximum power outtake for each driveline type, a battery system state-of charge (SOC), a fuel tank level, a maximum noise level, and a driveline type specific start-up procedure. This allows further refinement of the control strategy, which is an advantage.

There is also disclosed herein control units, vessels, computer programs, computer readable media, and computer program products associated with the above discussed advantages.

Generally, all terms used in the claims are to be interpreted according to their ordinary meaning in the technical field, unless explicitly defined otherwise herein. All references to "a/an/the element, apparatus, component, means, step, etc." are to be interpreted openly as referring to at least one instance of the element, apparatus, component, means, step, etc., unless explicitly stated otherwise. The steps of any method disclosed herein do not have to be performed in the exact order disclosed, unless explicitly stated. Further features of, and advantages with, the present invention will become apparent when studying the appended claims and the following description. The skilled person realizes that different features of the present invention may be combined to create embodiments other than those described in the following, without departing from the scope of the present invention.

BRIEF DESCRIPTION OF THE DRAWINGS

The above, as well as additional objects, features and advantages, will be better understood through the following illustrative and non-limiting detailed description of exemplary embodiments, wherein:

Figure 1 shows a marine vessel with a hybrid driveline arrangement;

Figure 2 illustrates a marine vessel hybrid driveline control strategy;

Figure 3 illustrates optional components in a driveline control system;

Figure 4 is a flow chart illustrating methods;

Figure 5 schematically illustrates a control unit; and

Figure 6 shows an example computer program product;

DETAILED DESCRIPTION

The present disclosure will now be described more fully hereinafter with reference to the accompanying drawings, in which exemplary embodiments are shown. The disclosure may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided for thoroughness and completeness. Like reference character refer to like elements throughout the description.

Figure 1 illustrates an example marine vessel 100 comprising a hybrid driveline 110 and a control unit 180 arranged to control the hybrid driveline 110 according to a control strategy. The control unit 180 is optionally connected via wireless link to one or more online resources, such as a server 190, which may form part of a cloud-based system of information sources. The server may also form part of the information processing system of the vessel 100, i.e. , the server 190 may in some cases assist the control unit 180 to perform computation tasks.

The hybrid driveline 110 comprises two or more driveline types connected to propulsion means 120. In other words, the vessel 100 implements a propulsion system comprising two or more driveline types. The propulsion means can, for instance, be a shared propeller to which more than one driveline is connected, but each driveline can also have its own propulsion means, i.e., a combustionengine based driveline can be connected to one propeller axle while an electric driveline is connected to another propeller axle. The techniques disclosed herein are applicable with both parallel and series hybrid systems. A combustion engine need not be directly connected to any driveline since it may also be used in a genset or the like to charge electric batteries of an electric driveline. More than two different driveline types can also be used in some types of marine vessels.

Each driveline type comprises a power source 130, 150 and an energy source 140, 160. A combustion engine driveline, for instance, comprises a combustion engine power source arranged to draw fuel from a fuel tank energy source. An electric drive comprises an electric machine power source arranged to draw electric current from an electrical energy source, such as a battery or a supercapacitor. Some examples of driveline types to which the teachings herein are applicable comprise diesel-based drivelines, gasoline-based drivelines, fuel cell electric systems, hydrogen-based combustion drivelines, battery-powered electric drivelines, a liquefied natural gas, LNG, based drivelines.

The amount of energy available to the different driveline types on a vessel may differ significantly. An electric driveline powered by some form of electrical energy storage system often has a significantly smaller amount of available energy compared to a combustion engine powered from a fuel tank. Thus, in a vessel with a first driveline type and a second driveline type, it is common that the first driveline type has much less available energy compared to the second driveline type, e.g., on the order of 10% or less.

The opportunity for replenishing the energy supply of the different driveline types may also differ. For instance, a given route may comprise one or more electrical charging stations where the electrical energy store can be replenished, but no opportunities for filling up the fuel tanks of the combustion engine driveline type.

A key aspect in the control of a marine vessel comprising a hybrid driveline is the power outtake balance between the driveline types along a given route. Obviously, a control strategy in which all power is always drawn from a dieselbased driveline and no power from the electric driveline will not be associated with any significant reduction in CO2 emission, while a control strategy which only uses the electric driveline will most likely not be able to get the marine vessel from the start of the route to the end, at least for longer routes.

It is difficult to predict energy consumption by a marine vessel with sufficient accuracy for power planning, i.e. , to determine how and when to use different driveline types in a hybrid driveline system. Known methods of predicting energy consumption in a marine vessel are typically based on weather forecasts, currents, and other environmental parameters. The known methods of energy prediction show satisfactory results when used to predict energy consumption in larger vessel for longer high sea transport operations, but the predictions obtained using known methods are not helpful for short-sea costal transport by smaller ships due to the low accuracy in the predictions.

Many marine applications are either repetitive (the ship travels back and forth along the same route many times in a row) like a ferry operation or semi- repeatable where the ship performs similar transport missions in a repetitive manner. These repetitive behaviors together with advanced self-learning systems can be used for more accurate energy planning suitable for smaller vessels in coastal traffic. For instance, suppose that a ferry travels from A to B and back again over and over again. Suppose also that the ferry has a hybrid driveline with a Diesel engine and an electric machine. It may be difficult to estimate the distance possible to cover using the electric machine. However, since the vessel travels along the same route over and over again, a model of range can be iteratively refined, resulting in a very accurate prediction of range for the electric machine along the route. This model is then specifically developed for the vessel and the driveline, travelling along the route, resulting in an increase in prediction accuracy. This way the control strategy for managing the propulsion system of a vessel that travels along a route in a repetitive manner can be improved compared to known methods.

To summarize, the present disclosure builds on the idea that the operation of marine vessels, i.e. , the routes travelled and the operating conditions along the routes encountered, are known beforehand or at least possible to predict a- priori. This provides a unique opportunity to adjust the balance between different types of drivelines to account for the conditions of a particular route in an iterative manner, having regard not only to the state of the vessel drivelines, but also to the operating conditions which will be encountered along the route. For example, a vessel 100 used to service a ferry route over a known route from one location to another location over and over again can collect data over time which indicates, e.g., a required power to complete one pass of the route, as function of environmental conditions (wind, wave height, sea current, temperature, etc.). This collected data can be used to configure a model of the vehicle hybrid driveline. The model is then based on the environment conditions and will provide information on a predicted power requirement for the route given a set of input environment conditions, which can be obtained from a recent weather forecast report. The control unit can then optimize the control strategy for the hybrid driveline of the vessel, based on the model, in an iterative manner as the vessel traverses the route or a similar route over and over again. The optimization criteria can be, e.g., CO2 emission for the route, total operating cost and/or long-term maintenance cost. In case the current weather conditions have not previously been experienced, then an extrapolation can be performed using similar environment conditions that have been encountered by the vessel previously.

Boundary conditions of the optimization algorithm can be defined for determining the control strategy. The boundaries can be used to ensure not only that the power output is met but also target time and distance are allowing the vessel to complete the trip within expectations. The boundaries will also allow for different safety margins dependent on the operation and user expectations. Energy source left at arrival for example will be a factor that is countering the cost effective solution but could be used because of upcoming trips or safety margins.

The techniques discussed herein are most advantageously used in workboats such as passenger ferries, cargo ferries, tugboats, pilot boats, crew transfer vessels (CTV) and tankers, but may generally be applied in any type of marine vessel.

Some aspects of the present disclosure relate to a computer implemented method for determining a marine vessel 100 hybrid driveline 110 control strategy 240, where the hybrid driveline 110 of the vessel propulsion system comprises at least a first driveline type 130, 140 and a second driveline type 150, 160 different from the first driveline type, such as an electric part and a combustion-engine based part. The control strategy comprises a power outtake P1 , P2 from the first driveline type 130, 140 and from the second driveline type 150, 160 along a route 210 to be traversed by the marine vessel 100 in a repetitive manner, and where the control strategy 240 has an associated emission target. The method comprises determining a total power requirement Ptotal for the marine vessel 100 along the route 210 to be traversed, at least in part by monitoring consumed power as the vessel repetitively traverses the route. In other words, the repetitive traversing of the same route allows the system to “sample” the energy consumption, given different operating conditions along the same route. The method also comprises obtaining power- to-emission relationship data for each of the driveline types each time the vessel traverses the route, where a power-to-emission relationship is indicative of an emission by the driveline type when operating at a given power outtake P1 , P2. The method furthermore comprises determining the marine vessel hybrid driveline control strategy 240 based on the total power requirement Ptotal for the route and on the power-to-emission relationships, such that the sum of power outtakes from the driveline types is greater or equal to the total power requirement Ptotal along the route 210 and such that the sum of emissions from the driveline types meets the emission target.

Figure 2 illustrates the concept of a hybrid driveline control strategy 200. A marine vessel traverses a route 210 from a start location 21 1 to a destination 212 at some distance D from the start. The route 210 is normally not straight, in particular if the route is a coastal route. Operating conditions along the route, such as wind 250, wave height 255, and current 260, are also likely to differ. The operating conditions may differ from one day to another for the entire route (the wind along the entire route may, e.g., be strong or weak), or locally (such as occurrence of strong sea current along some sub-section of the route). However, a vessel travelling the route over and over again will be able to map, e.g., weather conditions to power requirements and emissions along the route. Thus, a vessel travelling along a given route in a repetitive manner will be able to develop an accurate model of energy consumption and emission by the different driveline types over time. Thus, the control strategy can be refined over time for improved accuracy compared to known methods.

The control strategy 240 for controlling the driveline 110 of the marine vessel 100 comprises a power outtake P1 , P2 from the first driveline type 130, 140 and from the second driveline type 150, 160 along the route 210. This means that, at each position along the route, the control strategy determines how much power to draw from the first driveline type and from the second driveline type. Note that the hybrid system can be both a parallel system and a series system. A combustion engine can, for instance, be used to power a propeller or in a genset to charge the batteries of an electric driveline. The control strategies discussed herein are also associated with some form of emission target, such as a pre-determined emission amount for the route 210, or a minimization of emission along at least a part of the route 210.

With reference also to the flow chart in Figure 4, the method proposed herein comprises determining S1 a total power requirement Ptotal for the marine vessel 100 along the route 210 to be traversed. This total power requirement basically indicates how much power that needs to be generated by the hybrid driveline of the vessel in order to complete the route. This total power requirement Ptotal if generally a function of the length of the route and can also be a function of the environment conditions. A strong sea current can for instance have a large effect on power requirement for a given route, and so can wind and wave conditions along the route.

The method also comprises obtaining S2 a power-to-emission relationship for each of the driveline types. A power-to-emission relationship is indicative of an emission by the driveline type when operating at a given power outtake P1 , P2. Thus, for each driveline type, a mapping between power outtake and emission is available. This means that the system can determine an amount of emission for a given blend of the driveline types.

The method furthermore comprises determining S3 the marine vessel hybrid driveline control strategy 240 based on the total power requirement Ptotal for the route and on the power-to-emission relationships, such that the sum of power outtakes from the driveline types is greater or equal to the total power requirement Ptotal along the route 210 and such that the sum of emissions from the driveline types meets the emission target.

The graph at the bottom of Figure 2 shows an example control strategy 240. A power requirement for the route to be traversed by the vessel has four different levels along the route; 270, 271 , 272 and 273. Initially the power requirement is rather low, then increases, before again returning to a low level. The power outtake P1 increases when power requirement is high, while the power outtake P2 is kept rather constant. The power outtake P2 could perhaps be from an electric driveline which is not very efficient at very high power outtakes, while the power outtake P1 could represent the use of a combustion engine.

Figure 3 illustrates an example system 300 where the techniques discussed herein can be implemented. The system 300 comprises a number of optional input data sources 330-335, and generates one or more output signals 340, such as a rudder control signal 360, control signals for the vessel driveline 360, and also control signals for one or more auxiliary systems on the vessel, such as power take-outs and various forms of on-board equipment. The system 300 may also be configured to receive manual input 390, i.e., various forms of configuration data and also direct controls. A shore connection 395 may also be managed by the system 300. The processing circuitry 510 and the storage device 520 will be discussed in more detail below in connection to Figure 5.

Thus, it is understood that the system 300 gathers information from different sources, which may vary between systems and also over time. One source of information may be useful for some inference tasks while another may be more useful for other tasks. Some of the systems disclosed herein also comprise data source selection mechanisms. These mechanisms are realized by training using different types of data. The system can also be configured to detect which data sources that are available, and perform the tasks based on the available information. The primary sources of information are often GPS, compass data, and powertrain energy consumption data, i.e., fuel consumption or electrical energy consumption. Depending on the vessel, one or more secondary sources of information might be also available; for example, sonar, wind sensors or speed through water (sea current measurement sensor), as well as data indicating use of interceptors and/or water jets. Sensors measuring the setup of the vessel can also be used, such as trim setup and load balancing, also factors like passengers onboard or cargo mass can be included in the models. Online information, e.g., received from the server 190 can be also used in parallel including: plotter information, wind data, sea current data, wave height data from weather forecast datasets. Energy consumption will be recorded by measuring different energy sources like main engines and axillary power. Rudder position can be used as input to the system 300 as well. Shore connection information (energy input from land) can be used to identify charging capacity and charging time.

Automatic Identification System (AIS) information shows other vessels in the same area. According to some aspects, the system 300 receives data input also from these other vessels, which can be used in the system 300. This data can, for instance, be received via the server 190.

The data obtained by the system 300 is stored and processed in the processor unit 510. Data can also be sent to the server 190 via the wireless link and processed further, e.g., using a connectivity solution like Wi-Fi, satellite connection or 4G/5G cellular access.

The system 300 generally relies on a model of the vessel 100, which describes, e.g., its CO2 emission, running cost and long-term maintenance cost as function of the input data 330 comprising environment state variables such as wind, current, wave height, and the like. This model can be realized as a low complexity look-up table which describes power requirements for given routes and emission data for given driveline control settings. This look-up table can be populated by data collected from previous operations along the same or a similar route. The model can also be more advanced, such as a neural network or the like, trained using data obtained from previous operations by the same vessel 100 or by a similar vessel. The server 190 may collect data from one or more vessels and subsequently train the model based on the obtained data.

The route 210 to be traversed by the marine vessel 100 may be an a-priori known route. A-priori known routes are routes which extend from one location to another location via a predetermined path (such as a ferry line). Known routes may have been repeated enough in the past to enable creating a very accurate model for energy consumption for different environmental conditions, e.g., different wind, current and wave conditions. When the route is known and well documented in terms of the impact of environment conditions, energy consumption is also at least approximately known for different speed profiles given the different environmental conditions. These states can be presented as a look-up table or using a model. The model can be, e.g., a statistical model or a trained artificial intelligence structure like a neural network. The outcome after the model has been generated can be a sort of look-up table. Then, if the current operation conditions that are experienced by the vessel do not exist in the model, an extrapolation can be performed.

Herein, repeatable generally means that enough data from past operations by the vessel 100 or by similar vessels is available, which enables predictions of operation performance given various environment conditions for different routes. The route 210 to be traversed by the marine vessel 100 may also be a route determined by a route planning method, based on a known start location and destination. This route is not deterministically known but can be inferred from a known start and destination. A user may for instance input a destination, such as a home harbor location, and the system then proposes a route from the current location of the vessel 100 to the desired destination. This route is not deterministically known, but still often very accurate. The route to be travelled can be identified either by manual selection by user from a list of proposed routes, or by a separate model defining the most possible destination based on time, current location, and other relevant parameters. The user can of course have the option to override the route identification.

The route suggested by the path planning tool can be determined based on minimization of a cost function, e.g., using path planning algorithm like the Dijkstra algorithm or some form of rapidly exploring random tree (RRT) algorithm. The route planning is preferably updated if the vessel starts to deviate significantly from the preplanned path.

The route 210 to be traversed by the marine vessel 100 may be a route determined at least partly based on a current location of the marine vessel 100, on a prediction of destination based on previous routes traversed by the marine vessel 100, and on a route planning method based on current location and predicted destination. The prediction can also be based on a date and/or time of day. The prediction can also be based on a route travelled by the marine vessel 100 up to the current location.

An Al model can be trained based on a previously acquired dataset (GPS, time, and date), gathered at the model creation stage. The model can be expanded based on other factors like user profile to personalize the selection. The system continues to choose the most probable destination if there is no user overriding or by automatic deviation detection from the system. If the system has identified significant deviations from an assumed route, then a new destination can be estimated and selected by the model.

In case the destination of the route changes, then the control strategy may need to be updated S4. If the destination changes either by detecting a deviation (outside tolerance level of the model) or by user input, then a new control strategy will be selected.

The route 210 to be traversed by the marine vessel 100 may be associated with one or more environment conditions of one or more route segments, such as wave height, current, and wind along the route 210. If such route conditions are known, then they can be considered in the determination of the control strategy.

Inputs of the model may comprise wind direction, wind speed, sea current strength and direction, as well as wave height and direction obtained either from onboard sensors or from online weather databases. When the state of the sea is known for different locations from start to end of the trip an energy model can be created based on either look-up tables or a statistical/AI model. These factors will be presented as overriding factors that will increase the fuel consumption from baseline where the baseline is calm sea.

The method may, according to an example, comprise determining S11 the total power requirement Ptotal at least in part based on a target speed profile of the marine vessel 100 along the route 210. The desired speed to be maintained along the route of course influences the total power requirement. The faster the vessel travels the more power is required. The relationship between speed and power requirement is often complex, and seldom linear. The model is preferably configured to reflect this dependency between desired speed and power requirements. When the destination and time is known, a power requirement can be predicted for every location in the travel route considering the environmental parameters. During the operation, the system will determine the energy need based on the remaining distance and time. This might lead to a recommendation of higher or lower speed, hence modifying the control strategy.

The method may, according to an example, comprise determining S12 the total power requirement Ptotal at least in part based on a target noise level of the hybrid driveline 110 along the route 210. If the vessel is passing certain areas where a maximum noise level is required, either by law or best practices, then noise will be considered as a constraint in determining a control strategy e.g., by relying only on low noise driveline like electric.

The power-to-emission relationship can at least partly be based on an obtained efficiency metric E1 , E2 for each driveline type and on respective constant conversion factors K1 , K2 for each driveline type, S21 .

Emission per unit of power can be calculated based on emission map from the supplier or by measuring the fuel consumption of the driveline. For electricity, emission can be calculated based on local source of energy.

The method may also comprise determining S31 the marine vessel hybrid driveline control strategy 240 as a solution to an optimization problem with an objective function formulated in terms of emission for a given power outtake distribution along the route 210. The optimization can be based on various metrics, such as CO2 emission, running cost and long-term maintenance cost.

The optimization problem can of course also be solved by obtaining S311 the solution to the optimization problem from a machine learning (ML) algorithm, and/or by obtaining S312 the solution to the optimization problem from evaluating a pre-determined set of candidate control strategies.

According to another example, the method may comprise determining S32 the marine vessel hybrid driveline control strategy 240 based on a pre-determined set of candidate rules.

The method may also comprise determining S33 the marine vessel hybrid driveline control strategy 240 based on one or more constraints, where the constraints comprise any of: a maximum power outtake for each driveline type, a battery system state-of charge (SOC) a fuel tank level, a maximum noise level, and a driveline type specific start-up procedure.

The method may furthermore comprise determining S34 the marine vessel hybrid driveline control strategy 240 based on information related to energy replenishment locations along the route 210, e.g., whether there are charging stations for charging the electrical energy storage of the vessel available along the route 210 or not.

Charging stations along the route are used for a time that optimizes the cost of the travel in total. Longer stops for charging would increase travel time but allow for a better cost in terms of measurements, like CO2, noise, etc., this can then be optimized according to the active cost function.

Figure 5 schematically illustrates, in terms of a number of functional units, the components of a control unit 500 according to embodiments of the discussions herein. Processing circuitry 510 is provided using any combination of one or more of a suitable central processing unit CPU, multiprocessor, microcontroller, digital signal processor DSP, etc., capable of executing software instructions stored in a computer program product, e.g., in the form of a storage medium 530. The processing circuitry 510 may further be provided as at least one application specific integrated circuit ASIC, or field programmable gate array FPGA. Particularly, the processing circuitry 510 is configured to cause the control unit 500 to perform a set of operations, or steps, such as the methods discussed in connection to Figure 4 and generally herein. For example, the storage medium 530 may store the set of operations, and the processing circuitry 510 may be configured to retrieve the set of operations from the storage medium 530 to cause the control unit 500 to perform the set of operations. The set of operations may be provided as a set of executable instructions. Thus, the processing circuitry 510 is thereby arranged to execute methods as herein disclosed.

The storage medium 530 may also comprise persistent storage, which, for example, can be any single one or combination of magnetic memory, optical memory, solid state memory or even remotely mounted memory.

The control unit 500 may further comprise an interface 520 for communications with at least one external device. As such the interface 520 may comprise one or more transmitters and receivers, comprising analogue and digital components and a suitable number of ports for wireline or wireless communication.

The processing circuitry 510 controls the general operation of the control unit 500, e.g., by sending data and control signals to the interface 520 and the storage medium 530, by receiving data and reports from the interface 520, and by retrieving data and instructions from the storage medium 530. Other components, as well as the related functionality, of the control node are omitted in order not to obscure the concepts presented herein.

Figure 6 illustrates a computer readable medium 610 carrying a computer program comprising program code means 620 for performing the methods illustrated in Figure 4 and the techniques discussed herein, when said program product is run on a computer. The computer readable medium and the code means may together form a computer program product 600.