Marine vessel power systems

Abstract

Disclosed is a computer-implemented method for determining a configuration for a vessel power system. The vessel power system comprises a plurality of power system components including a plurality of power generators, each power generator configured to generate a type of power, and at least one propulsor to cause the vessel to move, each propulsor configured to be powered by power from at least one power generator. The power system is operable to cause a type of power to be delivered from a power generator to a propulsor to drive the propulsor. The power system is operable in a plurality of machine modes, each machine mode corresponding to respective configurations of the power system components such that a type of power is delivered from at least one power generator to at least one propulsor. The method comprises obtaining power demand data indicating a current power demand of a marine vessel to achieve a vessel objective, and determining, for each one of a plurality of machine modes and for a plurality of power output configurations for each machine mode, a specific fuel oil consumption for the vessel to achieve the objective and for the vessel power system to meet the power demand. Each power output configuration for each machine mode corresponds to respective power output levels of each power generator configured to deliver power to a propulsor in that machine mode, the sum of the power output levels for each power generator in each power output configuration being equal to the current power demand.

Description

[0001] MARINE VESSEL POWER SYSTEMS

[0002] The present disclosure relates to configurations fora power system of a marine vessel, and to processes for configuring such power systems.

[0003] BACKGROUND

[0004] In general, there are various factors to consider when determining the configuration of a vessel power system.

[0005] US 11598282 B1 relates to systems and methods for optimizing vessel fuel consumption. According to US282, environmental data and vessel operational data are collected using sensors onboard the vessel and input to a machine learning model that is trained to generate a predicted required power for a given voyage. The predicted required power is input into an engine-specific machine learning model for each thrust engine on board the vessel running in a particular engine configuration, the engine-specific machine learning model being configured to output a predicted fuel consumption for a specific thrust engine. A predicted fuel consumption for the engine configuration may be determined as the sum of the engine-specific predicted fuel consumption determined for each running thrust engine. Then, the engine configuration with the lowest total predicted fuel consumption may be selected as the optimum energy configuration and the vessel may be controlled according to that selected optimum energy configuration.

[0006] SUMMARY OF THE INVENTION

[0007] According to this disclosure there is provided a way of determining a configuration for a power system (or power supply unit) of a vessel, for example a marine vessel. The vessel power system may comprise a number of power generators configured to generate a type of power. The vessel power system may comprise a number of propulsors configured to be driven be power generated by at least one power generator. A propulsor being powered by a type of power is associated with a “machine mode” of the power system in that powering the vessel according to a particular machine mode (e.g. of a plurality of possible machine modes) corresponds to powering the vessel using a type power generated by a power generator. For each machine mode there may be a plurality of power output configurations corresponding to respective power output levels of the power generators operable in that machine mode. By way of example, a machine mode may be diesel-mechanic mode where a plurality of main engines are operable to deliver mechanical power to drive a propulsor, and a plurality of power output configurations for the main engines may be possible, each power output configuration corresponding to respective power output levels for the main engines to meet a power demand.There may therefore be a high degree of freedom regarding the choice of how a vessel is powered.

[0008] According to this disclosure there is provided a system and method for determining a vessel power system configuration that is associated with a particular specific fuel oil consumption. The term “vessel power system configuration” and should also be considered synonymous with the vessel power system being operated in a particular machine mode according to a particular power output configuration. According to this disclosure there is provided a system and method for determining a specific fuel oil consumption for a plurality of power output configurations and for a plurality of machine modes, thereby enabling a vessel configuration to be determined in a robust way that considers a large variety of factors that could affect the vessel’s control.

[0009] Examples presented herein allow a marine vessel to be operated efficiently and environmentally conscientiously, having regard to environmental regulations becoming increasingly strict and the rising prices of fuel. Given that more attention is being paid to greenhouse gas emissions, and that vessels while operating in a transit phase (where they are travelling at a constant speed) may emit greenhouse gases, the present disclosure allows a vessel to be operated in its transit phase in an environmentally conscious way. Moreover, for many vessel operators, energy efficiency and the reduction of fuel consumption are important considerations and the present disclosure allows a vessel to be operated with these in mind. It will therefore be appreciated that the present disclosure allows a vessel to be operated in an optimised way (particularly in transit) by utilising the vessel power systems in an energy efficient way that reduces emissions and fuel oil consumption.

[0010] According to an example of the present disclosure there is provided a computer-implemented method for determining a configuration for a vessel power system, the vessel power system comprising a plurality of power system components including: a plurality of power generators, each power generator configured to generate a type of power, and at least one propulsorto cause the vessel to move, each propulsor configured to be powered by power from at least one power generator, the power system being operable to cause a type of power to be delivered from a power generator to a propulsor to drive the propulsor, the power system being operable in a plurality of machine modes, each machine mode corresponding to respective configurations of the power system components such that a type of power is delivered from at least one power generator to at least one propulsor, the method comprising: obtaining power demand data indicating a current power demand of a marine vessel to achieve a vessel objective, and determining, for each one of a plurality of machine modes and for a plurality of power output configurations for each machine mode, a specific fuel oil consumption for the vessel to achieve the objective and for the vessel power system to meet the power demand, wherein each power output configuration for each machine modecorresponds to respective power output levels of each power generator configured to deliver power to a propulsor in that machine mode, the sum of the power output levels for each power generator in each power output configuration being equal to the current power demand.

[0011] The method may further comprise: transmitting the machine mode and power output configuration that is associated with the lowest specific fuel oil consumption to a vessel power system control unit.

[0012] The method may further comprise: causing the power system to be operated in the machine mode and according to the power output configuration associated with the lower specific fuel oil consumption.

[0013] The method further comprise: obtaining vessel objective data indicating the objective that the vessel is to achieve.

[0014] The method may further comprise: obtain operational data indicating the machine modes in which the power system of the vessel can be configured to operate.

[0015] The objective that the vessel is to achieve may be associated with the vessel operating in a transit mode.

[0016] The method may further comprise: obtaining input indicating that the vessel is to be operating, or is operating, in a transit mode.

[0017] Determining the specific fuel oil consumption, for each one of a plurality of machine modes and for a plurality of power output configurations for each machine mode, may be based on a speed of the vessel.

[0018] The method may further comprise: obtaining data indicating the speed of the vessel. The method may further comprise: determining, for each machine mode and for each power output configuration, the specific fuel oil consumption for the vessel to achieve the objective and to meet the power demand for a plurality of different vessel speeds.

[0019] The method may further comprise: transmitting the machine mode and power output configuration and vessel speed that is associated with the lowest specific fuel oil consumption to a vessel power system control unit.

[0020] The method may further comprise: causing the power system to be operated in the machine mode and according to the power output configuration, to thereby cause the vessel to be operated at the associated vessel speed, that is associated with the lowest specific fuel oil consumption.

[0021] The method may further comprise: determining, for each machine mode and for each power output configuration, the specific fuel oil consumption by: causing the current power demand to be input into a machine learning model trained based on historic data relating to specific fuel oil consumptions for a plurality of power output configurations and for a plurality of machine modes to output at least one specific fuel oil consumption associated with operating the vessel power system in at least one machine mode according to at least onepower output configuration to meet the current power demand of the vessel based on receiving the current power demand of the vessel as input, and

[0022] obtaining the specific fuel oil consumption output by the machine learning model.

[0023] The method may further comprise: determining, for each machine mode and for each power output configuration, the specific fuel oil consumption by: causing the current power demand of the vessel to be input into a plurality of machine learning models, each machine learning model in the plurality being trained based on historic data relating to specific fuel oil consumptions for a plurality of power output configurations in a respective machine mode to output a plurality of specific fuel oil consumptions associated with operating the vessel power system in the respective machine mode according to the plurality of power output configurations to meet the current power demand of the vessel based on receiving the current power demand of the vessel as input, and obtaining the specific fuel oil consumption output of each machine learning model for each machine mode.

[0024] The or each machine learning model may be trained based on historic data relating to specific fuel oil consumptions for a plurality of power output configurations and for the or each machine mode for the vessel operating in a transit mode.

[0025] The plurality of machine modes may comprise at least one of: a diesel mechanic mode in which at least one power generator is configured to generate mechanical power to power at least one propulsor, a diesel electric mode in which at least one power generator is configured to generate electrical power to power at least one propulsor, and a hybrid mode in which at least one power generator is configured to generate mechanical power to power at least one propulsor at least one power generator is configured to generate electrical power to power at least one propulsor.

[0026] The power generators of the vessel power system may comprise one or more of: a main engine configured to generate mechanical power, an auxiliary engine configured to generate electrical power, a generator configured to generate electrical power, and a motor configured to generate mechanical power.

[0027] The method may further comprise: obtaining vessel operational data, wherein the specific fuel oil consumption is determined for each machine mode and for each power output configuration based on the vessel operational data.

[0028] The vessel operational data may comprise at least one of: information relating to at least one component of the vessel power system and the flow rate of fuel to at least one power generator.

[0029] The method may further comprise: obtaining data indicating at least one of: the vessel speed, an operational state of at least one component of the vessel power system, and the flow rate of fuel flow to at least one power system, determining a machine mode in which the vessel is operating, determining a number of power generators of the vessel power systemthat are running according to respective power output levels, determining the specific fuel oil consumption of the vessel, and storing the specific fuel oil consumption, the power output configuration being the respective power output levels of the power generators, a route or a plurality of routes the vessel could traverse, a weather condition or set of weather conditions the vessel could be subject to while traversing a given route, power demand data as described above, an emission (e.g. greenhouse-gas or carbon) emission for the vessel while traversing a given route, and the machine mode, as training data.

[0030] The power demand data may comprise an obtained weather forecast along a given route or plurality of routes. The power demand data may be obtained, or may comprise, an energy plan. The power demand data may be an obtained power demand based on a given route and on a weather forecast along that route and may comprise a power demand to operate the vessel on that route in weather conditions defined by the obtained weather forecast. Alternatively or additionally, the power demand data may comprise a total nonpropulsion load for the vessel, the total being the sum of individual non-propulsion loads for each non-propulsion load for the vessel (or each working or operating non-propulsion load for the vessel given for a route). The power demand for each non-propulsion load (and therefore for the total) may be based on whether the or each non-propulsion load is static or variable; seasonal; the operational mode of the vessel; and / or the time of day. The power demand data may therefore comprise a power demand to operate a vessel on a route in a given weather condition or set of weather conditions, and optionally taking into account the obtained power demand of each non-propulsion load (e.g. the combined power demand of the total non-propulsion load).

[0031] The method may be performed for a given route whose emissions is determined to be less than a predetermined emissions threshold or the method may be performed fora plurality of routes and the selected SFOC is associated with a route having the lowest associated emissions.

[0032] According to an example of this disclosure there is provided a machine-readable medium storing instructions that, when executed by processor, cause the processorto perform the method as described above.

[0033] According to an example of this disclosure there is provided a system for determining a configuration for a vessel power system, the vessel power system comprising a plurality of power system components including: a plurality of power generators, each power generator configured to generate a type of power, and at least one propulsor to cause the vessel to move, each propulsor configured to be powered by power from at least one power generator, the power system being operable to cause a type of power to be delivered from a power generator to a propulsor to drive the propulsor, the power system being operable in a plurality of machine modes, each machine mode corresponding to respective configurations of thepower system components such that a type of power is delivered from at least one power generator to at least one propulsor,

[0034] the system being configured to: obtain power demand data indicating a current power demand of a marine vessel to achieve a vessel objective, and determine, for each one of a plurality of machine modes and for a plurality of power output configurations for each machine mode, a specific fuel oil consumption for the vessel to achieve the objective and for the vessel power system to meet the power demand, wherein each power output configuration for each machine mode corresponds to respective power output levels of each power generator configured to deliver power to a propulsor in that machine mode, the sum of the power output levels for each power generator in each power output configuration being equal to the current power demand.

[0035] The system may be further configured to: transmit the machine mode and power output configuration that is associated with the lowest specific fuel oil consumption to a vessel power system control unit.

[0036] The system may be further configured to: cause the power system to be operated in the machine mode and according to the power output configuration associated with the lower specific fuel oil consumption.

[0037] The system may be further configured to: obtain vessel objective data indicating the objective that the vessel is to achieve.

[0038] The system may be further configured to: obtain operational data indicating the machine modes in which the power system of the vessel can be configured to operate.

[0039] The objective that the vessel is to achieve may be associated with the vessel operating in a transit mode.

[0040] The system may be further configured to: obtain input indicating that the vessel is to be operating, or is operating, in a transit mode.

[0041] Determining the specific fuel oil consumption, for each one of a plurality of machine modes and for a plurality of power output configurations for each machine mode, may be based on a speed of the vessel.

[0042] The system may be further configured to: obtain data indicating the speed of the vessel.

[0043] The system may be further configured to: determine, for each machine mode and for each power output configuration, the specific fuel oil consumption for the vessel to achieve the objective and to meet the power demand for a plurality of different vessel speeds.

[0044] The system may be further configured to: transmit the machine mode and power output configuration and vessel speed that is associated with the lowest specific fuel oil consumption to a vessel power system control unit.The system may be further configured to: cause the power system to be operated in the machine mode and according to the power output configuration, to thereby cause the vessel to be operated at the associated vessel speed, that is associated with the lowest specific fuel oil consumption.

[0045] The system may be further configured to: determine, for each machine mode and for each power output configuration, the specific fuel oil consumption by: causing the current power demand to be input into a machine learning model trained based on historic data relating to specific fuel oil consumptions for a plurality of power output configurations and for a plurality of machine modes to output at least one specific fuel oil consumption associated with operating the vessel power system in at least one machine mode according to at least one power output configuration to meet the current power demand of the vessel based on receiving the current power demand of the vessel as input, and obtaining the specific fuel oil consumption output by the machine learning model.

[0046] The system may be further configured to: determine, for each machine mode and for each power output configuration, the specific fuel oil consumption by: causing the current power demand of the vessel to be input into a plurality of machine learning models, each machine learning model in the plurality being trained based on historic data relating to specific fuel oil consumptions for a plurality of power output configurations in a respective machine mode to output a plurality of specific fuel oil consumptions associated with operating the vessel power system in the respective machine mode according to the plurality of power output configurations to meet the current power demand of the vessel based on receiving the current power demand of the vessel as input; and obtaining the specific fuel oil consumption output of each machine learning model for each machine mode.

[0047] The or each machine learning model may be trained based on historic data relating to specific fuel oil consumptions for a plurality of power output configurations and for the or each machine mode for the vessel operating in a transit mode.

[0048] The plurality of machine modes may comprise at least one of: a diesel mechanic mode in which at least one power generator is configured to generate mechanical power to power at least one propulsor, a diesel electric mode in which at least one power generator is configured to generate electrical power to power at least one propulsor, and a hybrid mode in which at least one power generator is configured to generate mechanical power to power at least one propulsor at least one power generator is configured to generate electrical power to power at least one propulsor.

[0049] The power generators of the vessel power system may comprise one or more of: a main engine configured to generate mechanical power, an auxiliary engine configured to generate electrical power, a generator configured to generate electrical power, and a motor configured to generate mechanical power.The system may be further configured to: obtain vessel operational data, wherein the specific fuel oil consumption is determined for each machine mode and for each power output configuration based on the vessel operational data.

[0050] The vessel operational data may comprise at least one of: information relating to at least one component of the vessel power system and the flow rate of fuel to at least one power generator.

[0051] The system may be further configured to: obtain data indicating at least one of: the vessel speed, an operational state of at least one component of the vessel power system, and the flow rate of fuel flow to at least one power system, determine a machine mode in which the vessel is operating, determine a number of power generators of the vessel power system that are running according to respective power output levels, determine the specific fuel oil consumption of the vessel, and store the specific fuel oil consumption, the power output configuration being the respective power output levels of the power generators, a route or a plurality of routes the vessel could traverse, a weather condition or set of weather conditions the vessel could be subject to while traversing a given route, power demand data as described above, an emission (e.g. greenhouse-gas or carbon) emission for the vessel while traversing a given route, and the machine mode, as training data.

[0052] The power demand data may comprise an obtained weather forecast along a given route or plurality of routes. The power demand data may be obtained, or may comprise, an energy plan. The power demand data may be an obtained power demand based on a given route and on a weather forecast along that route and may comprise a power demand to operate the vessel on that route in weather conditions defined by the obtained weather forecast. Alternatively or additionally, the power demand data may comprise a total nonpropulsion load for the vessel, the total being the sum of individual non-propulsion loads for each non-propulsion load for the vessel (or each working or operating non-propulsion load for the vessel given for a route). The power demand for each non-propulsion load (and therefore for the total) may be based on whether the or each non-propulsion load is static or variable; seasonal; the operational mode of the vessel; and / or the time of day. The power demand data may therefore comprise a power demand to operate a vessel on a route in a given weather condition or set of weather conditions, and optionally taking into account the obtained power demand of each non-propulsion load (e.g. the combined power demand of the total non-propulsion load).

[0053] The system may perform the above-described processes for a given route whose emissions is determined to be less than a predetermined emissions threshold or the method may be performed for a plurality of routes and the selected SFOC is associated with a route having the lowest associated emissions.According to an example of this disclosure there is provided a cloud computing environment system comprising the system as described above.According to an example of this disclosure there is provided a computer hardware configured to implement the method, the system, or the cloud computing environment system as described above.

[0054] US 11598282B1 uses a predicted power (predicted by a first machine learning model) to determine an engine configuration, which may lead to inaccuracies as the predicted power may not reflect the actual power being used by the vessel. Moreover, that the power is predicted means that the US282 systems and methods may have limited, accurate, real-time applicability. Furthermore, in US282 to determine the predicted fuel consumption, a machine learning model is used for each thrust engine (the machine learning models are “engine specific”), and therefore individual engine fuel consumptions are determined for a candidate engine configuration. As a consequence, to arrive at a total predicted fuel consumption, the individual outputs from each engine-specific machine model are summed per engine configuration. As a further consequence, the process must be run again per engine configuration. In other words, to compare two engine configurations the engine specific machine learning models are used twice per thrust engine that is running in each configuration, and the results summed. This can easily lead to high computational overheads particularly when a large amount of engine configurations are to be considered. Finally, US282 does not take into account that modern vessels may have different machine modes and operational configurations, and modern vessels may therefore afford greater degrees of freedom fortheir operational control than the relatively simplistic vessels with which US282 is concerned.

[0055] One object of the present disclosure is to address these issues.

[0056] BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Examples of the present disclosure will be described in detail with reference to the accompanying drawings, which should not be considered limiting, in which:

[0058] Figure 1 shows a schematic diagram of a vessel in combination with a controller; Figures 2a and 2b show a schematic diagrams of vessel power systems;

[0059] Figures 3 and 4 show flowcharts illustrating example methods;

[0060] Figures 5-7 show schematic diagrams of machine learning models;

[0061] Figure 8 shows a flowchart of an example method for determining training data for training a machine learning model;

[0062] Figure 9 shows a schematic diagram of a process of deriving a machine learning model;

[0063] Figure 10 shows a schematic diagram of an example machine-readable medium in association with a processor;

[0064] Figure 11 shows a schematic diagram of an example system; and

[0065] Figure 12 shows a schematic diagram of an example cloud environment.DETAILED DESCRIPTION

[0066] These drawings should not be considered limiting, rather they are used for explaining and understanding the present disclosure.

[0067] Figure 1 shows a vessel 1 , depicted as a marine vessel, involved in a transit operation in which the vessel is moving (e.g. toward a destination, such as for the purpose of achieving a vessel objective or completing a vessel mission). The vessel 1 comprises a controller 2 which is an on-board or on-vessel controller. Equally, a controller 3 is depicted as being not on-board the vessel, e.g. remote from the vessel, in the Figure 1 example it is depicted on land (as opposed to the water) but the controller 3 could be in any location remote from the vessel, including at a location at sea. The on-board controller 2 and / or the remote controller 3 may be configured to determine a configuration for a vessel power system according to the processes described herein.

[0068] The vessel 1 comprises such a vessel power system comprising a plurality of power generators each power generator configured to generate a type of power (e.g. mechanical or electrical) that is to be used to drive at least one propulsor of the vessel power system to propel the vessel (e.g. in the transit mode) as will now be described.

[0069] Figure 2a schematically shows such a power system 200 for a vessel. The power system 200 comprises a number of components that are capable of being active or inactive (e.g. ON or OFF) depending on the type of power that the vessel is to use for its propulsion, which corresponds to the type of machine mode of the power system 200.

[0070] Figure 2 illustrates a power system capable of operating in a diesel-mechanic mode, a diesel-electric mode, or a hybrid mode. It will be appreciated therefore, that the power system 200 is capable of being operated according to a plurality of machine modes. Other power systems may be capable of operating in one mode. Nevertheless, the present disclosure applies to any power system for any vessel.

[0071] Elements 201a and 202a denote a propulsor of the vessel being connected to respective shafts 201 , 202 that are configured to drive the propulsors. The propulsors are depicted as propellers in this example but they may be any other component, or any plurality of components, configured to propel the vessel (e.g. to cause it to move), and although two propulsors are depicted the system 200 may comprise any number of propulsors. In the system 200 one of the propellers may be provided in the port-direction of the vessel and the other propeller may be provided in the starboard-direction. Hence an example power system may comprise a plurality of components for driving at least one port propulsor and a plurality of components fordriving at least one starboard propulsor. In this way, the power system may comprise two identical or near-identical systems, a port system for driving the at least one port propulsor and a starboard system fordriving the at least one starboard propulsor. For ease ofexplanation propeller 201a and the components in the upper portion of Figure 2 may be referred to as the port power system and the propeller 202b and the components in the lower portion of Figure 2 may be referred to as the starboard power system.

[0072] The propellers 201a, 202a are connected to the respective shafts 201 , 202 such that rotation of the shafts 201 , 202 causes the propellers 201a, 202a to rotate. As indicated in Figure 2 the shafts 201 , 202 are connected to respective gear systems 203, 204 such that rotation of the gear systems 203, 204 causes rotation of the shafts 201 , 202 which, in turn causes the propellers 201 a, 202a to rotate which moves the vessel and in this way, the power system 200 causes the vessel to move.

[0073] The gear systems 203, 204 may be caused to rotate by different types of power, which correspond to the machine mode in which the power system 200 is configured to operate. For example, the gear systems 203, 204 may be caused to rotate by mechanical power, electrical power and / or a combination of mechanical and electric power (e.g. hybrid power). To achieve this, the gear systems 203, 204 are respectively connected to main engines 205, 206. Each one of the main engines 205, 206 may be a diesel engine (e.g. an internal combustion engine configured to produce torque via the internal combustion of diesel fuel) or may burn heavy fuel oil. More specifically, the main engines (“MEs”) 205, 206 are each configured to cause the respective gear systems 203, 204 to rotate (e.g. an output shaft of a main engine may be connected to part of a respective gear system such as a shaft or a gear thereof). In this way, mechanical power produced by a main engine 205, 206 may drive one (or both) propellers 201a, 202a.

[0074] However, the power system 200 is configured such that the gear systems 203, 204 are configured to (alternatively or additionally) be driven by respective motors 213, 214. Each motor 213, 214 may be configured to produce mechanical power upon receiving electrical power as input. The motors 213, 214 may be refers to as “power take ins” (“PTIs”). The double lines in Figure 2 schematically illustrate mechanical connections between components whereas single lines are intended to illustrate electrical connections. As shown in Figure 2, each PTI 213, 214 is controlled by a respective drive unit 217, 218 (e.g. a variable frequency drive (VFD) or variable speed drive (VSD)) which is configured to control the speed of the respective PTI. The drive units 217, 218 may comprise an inverter or a converter depending on the type of drive unit that is used in the implementation.

[0075] It should therefore be appreciated that the gear systems 203, 204 may be exclusively driven as a result of power generated by a mechanical power generator in the form of one of the main engines 205, 206 or may be exclusively driven as a result of power generated be an electrical power generator (for example, one or more auxiliary engines to be described below) or the MEs may be used in conjunction with the PTIs whose purpose is to provide power tothe respective shafts to boost the power that is generated by the main engines 205, 206, e.g. they may provide temporary extra power (for example, extra power in a boost mode).

[0076] As also shown in Figure 2, each engine 205, 206 is connected to a respective generator 207, 208. Each generator 207, 208 is configured to receive mechanical power from a respective main engine (e.g. from an output shaft thereof) and configured convert the mechanical power into electrical power. In this way each generator 207, 208 is configured to generate electrical power and which may be used as an additional power source. The generators 207, 208 may also be referred to as “power take outs” (“PTOs”). As shown in Figure 2 each generator 207, 208 is connected (e.g. electrically connected) to an electrical bus bar 215 (e.g. via a wired connection). The busbar 215, which may also be referred to as a main switchboard, is configured to receive electrical power (e.g. electricity / electrical current) from the or each generator 207, 208 and is configured to transport and / or distribute electrical power (e.g. to another part of the vessel).

[0077] The power system 200 comprises four auxiliary engines (“AUX”), two associated with each propeller (209 and 211 for propeller 201 a and 210 and 212 for propeller 202a). Whereas the MEs are configured to directly drive the gear systems to drive the vessel’s propellers, the auxiliary engines are configured to generate electrical power for other systems aboard the vessel such as for one or more vessel systems, such as lighting, communication, navigation, pumps, HVAC systems or other equipment such as cranes, winches, and other machinery. The auxiliary engines are configured to generate electrical power (e.g. electricity / electrical current) and to supply it to the busbar, or circuit board, 215. Via the busbar 215, these other systems (lighting, communication systems etc.) may be supplied with electricity to be powered.

[0078] As stated above, the power system 200 is operable in a number of “machine modes.” The machine modes may be any one of a diesel mechanic (“DM”) mode, a diesel electric mode (“DE”), and a hybrid mode. Each one of these modes corresponds to a type, or configuration, of a powertrain of the vessel power system 200, the powertrain being the number components (e.g. engine components) of the power system 200 that are on and active for the power system 200 to deliver power to the propulsors 201 , 202 when operating in a given machine mode.

[0079] By way of example, when operating in a diesel-electric mode the power system 200 may be configured such that main engines 205, 206 and the generators 207, 208 are OFF and the PTIs 213, 214 are ON to power the propulsors. The auxiliary engines 209-212 may also be ON in this machine mode.

[0080] By way of another example, when operating in a diesel-mechanic mode the power system 200 may be configured such that at least one of the main engines 205, 206 are ON to power their respective propulsor(s).By way of a further example, when operating in a hybrid mode the power system 200 may be configured such that the generators 207, 208 are OFF and the main engines 205, 206, PTIs 213, 214, and the auxiliary engines 209-212 are be ON to power the propulsors.

[0081] Optionally, in each mode, electrical power generated by at least one generator a and / or at least one auxiliary engine may be delivered via the bus bar 215 to other loads that are connected to the electrical bus bar 215, which is configured to deliver power to other consumers on the vessel such as HVAC system(s), crane(s), winch(es) etc.

[0082] These examples are not exhaustive.

[0083] For example, the power system 200 may be configured in a diesel-mechanic “economy mode” in which one main engine is running to provide mechanical power to one propeller, while the other propeller is powered by electrical power supplied by the shaft generator. In such a machine mode, the main engine 205 may be ON to power the propeller 201a and the shaft generator 207 may be ON, the main engine 206 and shaft generator 208 may be OFF, the PTI 214 and the drive unit 218212 may be ON to deliver electrical power to the propeller 202a, the electrical bus being powered by the generator 207 and the propeller 202a being energized using the PTI 214 and the drive unit 218 in this mode.

[0084] The power system may additionally or alternatively be configured in a diesel-mechanic “normal mode” in which the main engines 205,206 is ON to power the propulsors and the shaft generators 207,208 are ON to power any other consumers connected to the switchboard 215.

[0085] The power system may additionally or alternatively be configured in a diesel-mechanic “speed mode” in which the shaft generators 207, 208 and the PTIs 213, 214 are OFF and the main engines 205, 206 and the auxiliary engines 209-212 are ON (the main engines being ON to power the propulsors and the auxiliary engines being ON to deliver power to any power consumers connected to the switchboard 215).

[0086] The power system may additionally or alternatively be configured in a diesel-mechanic “boost mode” in which all components are ON expect for the shaft generator 207 and 208. In this mode, the propellers are powered both from main engines 205 and 206 mechanically and from the PTI motors 213, 214, and drive units 217, 218 electrically using the auxiliary engines 209-212. By doing so, the propellers are able to be run at their maximum capacity in this mode.

[0087] Therefore, at least five machine modes are possible in the Figure 2 power system: DE, Hybrid, DMEconomy, DMNormal, DMSpeed, and DMBoost. It will be appreciated that each machine mode corresponds to a configuration of the power system, e.g. a number of components thereof, and that in each machine mode the power system is configured such that a type of power is generated by and / or delivered from at least one power generator (e.g. main engine, auxiliary engine, generator, and / or motor) to at least one propulsor, and thereforeeach machine mode is synonymous with a powertrain being formed to deliver that type of power from the generator to the propulsor.

[0088] Operation of the power system 200 may be under the control of controller, such as 2 or 3 described above).

[0089] When operating in each machine mode the power system is to deliver a power demand for the marine vessel so that it can achieve its objective (e.g. travelling at a constant speed). Therefore, in each machine mode where a plurality of power generators are operating to deliver a type of power to at least one propulsor, the sum of their respective power outputs are such that the power system can meet the power demand so the vessel can achieve its objective. The sum of the power output levels for each power generator is therefore equal to a current vessel power demand regardless of in which machine mode the power system is operating. However, it will be appreciated that different power output levels of the power generators may be possible for the power system to meet a given power demand. In other words, forthe power system to meet a given power demand while operating in a given machine mode, different power output configurations may be possible, with each power output configuration corresponding to respective power output levels of each generator per machine mode. It will furthermore be appreciated that each power output configuration corresponds to a specific fuel oil consumption (“SFOC”)

[0090] According to the present disclosure there is provided a process of determining a SFOC forthe power system operating according to a plurality of power output configurations and in a plurality of different machine modes. Said another way, according to the present disclosure there is provided a process of determining a SFOC for a plurality of different machine modes and a plurality of different power output configurations per machine mode. This allows the power output configuration and machine mode to be selected for which the SFOC is determined to be the lowest so that the vessel can be powered in an energy-efficient way.

[0091] The present disclosure is particularly applicable for a machine vessel operating in a “transit mode” where the vessel is to be travelling (e.g. from one waypoint to another), at least at a minimum speed (which may be referred to as a transit speed). During the transit mode, a certain power demand may need to be met. In other words, a certain power demand may be determined for the vessel to achieve and / or maintain the transit speed. The vessel power system may be operable in a plurality of machine modes, according to a number of power output configurations of the generators of each machine mode, for that power demand to be met and therefore for that transit speed to be met. Therefore, the present disclosure allows, for a given power demand (such as a power demand for the vessel to operate in the transit mode), a plurality of power output configurations to be determined for a plurality of machine modes such that the vessel’s power system achieves a total power output that is equal to the power demand (having regard to respective power output levels of the engine componentsthat are active in each mode). One power output configuration (corresponding to one machine mode) may then be selected (e.g. the power output configuration corresponding to the lowest SFOC) and the vessel power system may then be operated according to that selected configuration.

[0092] By way of a specific and non-limiting example, for a vessel to achieve a certain power demand, the vessel power system may operate in a diesel mechanic mode according to various power output configurations that correspond to different power outputs of each of the two main engines such that each power output configuration delivers the combined power output being equal to the power demand, and in a diesel electric mode according to various power output configurations that correspond to different power outputs of one or more auxiliary engines and PTIs such that each power output configuration delivers the combined power output being equal to the power demand. Say (purely for illustrative purposes) that six power output configurations in the diesel mechanic mode can deliver the power demand, and four power output configurations for the diesel electric mode can deliverthe power demand. In this illustrative example, two machine modes are possible according to which the vessel may be operated, and ten power output configurations are possible according to which the vessel may be operated (sixforthe DM mode, fourforthe DE mode). The disclosure herein may determine the SFOC for each of the two modes and power output configuration (e.g. each of the ten power output configurations) and one may then be selected according to which the vessel is operated.

[0093] The present disclosure presents a process (or method). As part of the process a “SFOC” may be determined for a plurality of machine modes and power output configurations per machine mode. This may be done “on-the-fly” e.g. on-board a vessel or off-board a vessel and will be referred to herein as “determining a SFOC”, e.g. ‘in-use’. In one example the SFOC may be determined using a machine learning (“ML”) model and, in these examples, this disclosure relates to deriving such a ML model. Firstly, we describe a process for determining the SFOC for different power output configurations and machine modes to operate a vessel. Secondly, we describe a process for deriving the ML model (according to which the SFOC may be determined).

[0094] Figure 2b schematically shows another power system 200 for a vessel. As for Figure 2a, the power system 200 is operable in a number of machine modes and comprises a number of components that are capable of being active or inactive (e.g. ON or OFF) depending on the type of power that the vessel is to use for its propulsion, which corresponds to the type of machine mode of the power system 200. As for Figure 2a, the Figure 2b power system comprises propulsors 201 , 202 each connected to respective main engines 205, 206. As for the MEs in Figure 2a, one or more MEs in Figure 2b may each be a diesel engine (e.g. an internal combustion engine configured to produce torque via the internal combustion of dieselfuel) or may burn heavy fuel oil. The MEs 205 may be configured to produce torque that is configured to rotate an output shaft. The propellers 201 , 202 (e.g. shafts thereof) are connected to the respective MEs 205, 206 (e.g. to shafts thereof) such that a respective ME 205, 206 causes (directly or indirectly) a propeller 201 , 202 to rotate.

[0095] As for Figure 2a, the power system of figure 2b comprises a plurality of components for driving at least one port propulsor and a plurality of components for driving at least one starboard propulsor, and in this way the power system may comprise two identical or nearidentical systems, a port system for driving the at least one port propulsor and a starboard system for driving the at least one starboard propulsor.

[0096] Unlike the Figure 2a power system which comprised a distinct PTI and PTO, the Figure 2b power system comprises “in-line” shaft generators 221 , 222 which each comprise a combined PTI / PTO. The combined shaft generators (or “PTI / PTOs”) 221 , 222 may each have the dual function of being able to use the rotation of a shaft connected to a respective propeller and ME to generate electricity and be powered by electricity to drive the shaft of a respective propeller 201 , 202 to thereby cause it to rotate. The power system 200 comprises a main switchboard 215, drive units 217, 218, and auxiliary engines 209-212 whose functions are as described above with respect to Figure 2a.

[0097] Each combined PTI / PTO 221 , 222 may comprise a rotor rotatably connected to a shaft (e.g. a propeller shaft) and that is rotatable within a stator. When an ME 205, 206 causes its own shaft to rotate (to thereby rotate the propeller shaft) rotation of the rotor within the stator of the PTI / PTO 221 , 222 may generate electricity. On the other hand, the stator may be energised by electricity, e.g. provided by an auxiliary engine 209-212 under the control of the main switchboard 215, to cause the rotor to rotate within the stator which causes the shaft to rotate which rotates the propeller 201 (and in this way the propeller may be rotated without using the main engine). Current through the stator may be controlled by the respective drive units 217, 218 (e.g. variable-frequency drive units).

[0098] It will therefore be appreciated that the Figure 2b power system is operable in a dieselmechanic mode in which at least one propeller 201 , 202 is driven by mechanical power from a main engine 205, 206 (and in which mode the PTI / PTOs 221 , 222 may be ON and operating to generate electricity to be supplied to the main switchboard 215 for other systems aboard the vehicle, or may be OFF; the auxiliary engines 209-212 may be off in this mode). It will also be appreciated that the Figure 2b power system is operable in a diesel-electric mode in which electrical power generated by at least one auxiliary engine 209-212 is converted to mechanical energy by the PTI / PTO 221 , 222 to rotate the propeller shafts (the MEs 205, 206 may be off in this mode). It will further be appreciated that the Figure 2b power system is operable in a hybrid mode in which hybrid power (e.g. using mechanical power from the MEs 205, 206 and electrical power from the auxiliary engines 209-212) is used to drive the propellers 201 , 202.As for Figure 2a, different power output configurations of the power generators may be possible in each mode to achieve a given power output demand, corresponding to different SFOCs for the power system to meet the power demand. Therefore, whilst Figures 2a and 2b illustrate different types of power systems, the processes described herein relate to determining machine modes and power output configurations (and the SFOCs) for any type of power system, including the type of system depicted in Figure 2a, the type of system depicted in Figure 2b, or indeed any other power system.

[0099] Determining a SFOC for a vessel (e.q. using a ML model)

[0100] Figure 3 shows a process 300 that may be performed by a controller (such as 2 or 3 as described above). The process may be to determine a SFOC for a vessel, e.g. for the vessel to operate in a transit mode.

[0101] At 302, the process 300 comprises obtaining power demand data indicating a current power demand of a marine vessel to achieve a vessel objective. At 304, the process comprises determining a specific fuel oil consumption (“SFOC”) for the vessel to achieve the objective and to meet the power demand. As indicated by the two looping arrows, block 304 is performed for each machine mode, and for each power output configuration per machine mode. As discussed above, a machine mode corresponds to a type of power generated by at least one power generator to power at least one propulsorof the power system for the vessel to achieve the vessel objective) and a power output configuration corresponds to the respective power output levels of each power generator that are active in a given machine mode. As stated above, the power output levels are to achieve a power demand which may depend on a vessel speed (e.g. a speed through water (“STW”) or speed over ground (“SOG”)).

[0102] An machine mode may be considered synonymous with a configuration of a powertrain within the vessel power system and may correspond to a number of active, or on, components of the power system. As the sum of the power outputs for each power output configuration is equal to the current power demand, block 304 determines a SFOC for a plurality of power output configurations for a plurality of machine modes such that the power demand can be met and such that the vessel can achieve its objective. As above, the power demand may be such that the vessel can achieve or maintain a speed (e.g. a STW or SOG, the SOG may be determined using GPS data).

[0103] The power demand may be associated with an amount of power required so that the vessel can achieve the vessel objective. In some examples the power demand may be determined (e.g. as opposed to estimated) and may be determined as the power required to operate the vessel in a mode, such as the transit mode (or transit configuration). Block 302 may comprise determining that the vessel is operating in transit. Additionally, or alternatively,block 302 may comprise determining the speed at which the vessel is travelling (e.g. a transit speed) which may be a STW or a SOG and may be determined by a controller from data received from a sensor. In these examples, block 302 may then comprise determining (e.g. calculating) the power required to continue to operate the vessel to achieve its objective (e.g. to arrive at a target destination) while maintaining that speed. In this way, some implementations of the process 300 may be performed once it is determined that the vessel is already operating in transit.

[0104] The vessel objective may therefore be associated with a mode of operation of the vessel, which may comprise a transit mode or any other mode. Accordingly, in one example, the power demand may be associated with an amount of power required so that the vessel can operate in transit mode, for example for the vessel to maintain a constant speed (e.g. a transit speed) in the transit mode, which may be any speed such as a STW or SOG. The vessel objective data may comprise further information (e.g. vessel heading, bearing, time to destination, an instruction to maintain speed etc.).

[0105] In one example, the power demand data may be based on an energy plan that is based on a route (e.g. a route that the vessel is to take or a candidate route that the vessel may take) and / or a weather forecast (either for a route that the vessel is to take or for a candidate route that the vessel may take). In these examples block 302 may comprise obtaining, as input, a target route for the vessel. As part of this input or as part of a separate input block 302 may comprise obtaining a weather condition estimate for that route (e.g. obtaining weather data comprising an estimate of at least one weather condition on the route). Block 302 may then comprise obtaining an energy plan for that route and this may be based on the weather condition estimate. In this way, the energy plan may provide the needed propulsion power along the route and the power demand data may be based on (e.g. may be part of) that energy plan. Block 302 may, alternatively, or additionally, comprise obtaining non-propulsion power demand data for the route or the relevant period of time. Non-propulsion power demand includes power demand ofthe vessel other than that required for the propulsion of the vessel, such as hotel loads and mission loads. Hotel loads typically includes ventilation, air conditioning, lighting, and various systems for passenger and operational needs. Mission loads may include power demand related to the missions of the vessel, such as deck machinery, cargo management.

[0106] The total non-propulsion load power demand data may comprise the sum of nonpropulsion load power demand data for a plurality of non-propulsion loads of the vessel. In some examples a non-propulsion load may be classified based on at least one of: whether the load is static or variable; whether the use ofthe non-propulsion load is seasonal (e.g. depends on the season); the operational mode of the vessel; or the time of day. It will be appreciated that the total non-propulsion load power demand may also be route-dependent and thereforemay be obtained either for a given route, or for a plurality of routes. In some examples, the power demand data may comprise the estimated total non-propulsion load and / or the propulsion power as determined based on the energy plan for at least one vessel route. It should also be appreciated that in some examples the energy plan may comprise the total non-propulsion load power demand data. This permits the integration of an “energy planner” sub-system that may be configured to use at least one planned route and weather forecast to estimate the propulsion demand along one or a plurality of routes, optionally additionally estimates the non-propulsion loads / base loads for the weather forecast for each nonpropulsion load along the or each route. Such a system may therefore provide a system implementing the process with an overview of the total power demand along one or a plurality of voyages which may be used by a system implementing the process to recommended different power system configurations to different states of the vessel (transit, DP, etc.).

[0107] It should be appreciated that this “energy planner” aspect may be performed again at a later time (e.g. repeated continuously at regular time intervals or periodically after a predetermined time period has elapsed) so that the process 300 can be continuously or periodically performed again and therefore can be adapted for if the route or weather forecast or non-propulsion load configuration changes overtime.

[0108] As indicated by the optional blocks 312-318 repeating the “energy planner” steps as described above (route and weather forecast data and optional non-propulsion load data) may be done in conjunction with any one or more of the blocks 312-318. E.g. any one or more of these parts of the process may be performed again when any one input changes or may be performed again continuously or periodically.

[0109] It will be appreciated that a route may comprise a number of segments and the energy plan and / or the non-propulsion load data (the total for a plurality of non-propulsion loads or individual load data for an individual non-propulsion load) may comprise (indeed may be the sum of) propulsion powerand load data forthe individual route segments, e.g. the combination of different route segments, that make up a given route).

[0110] Block 302 may also comprise obtaining emissions data which may be for a given route or for a plurality of routes if the process 300 is to be performed multiple times to compare SFOCs. Such emissions data may comprise data from an emissions sensor. Such emissions data may be based on the location of the vessel (which may be a predicted or estimated location of the vessel based on individual segments of a particular route) such as whether the vessel is currently located in or determined to be in, at a future time, an emissions control area, a zero emissions area, a narrow passage or any other area which could place emissionbased restrictions on the vessel. In this way, by obtaining emissions data, the SFOC may be route-based in that the process may be performed for a specific route that is based on emissions data (e.g. the route having the lowest associated emissions). Alternatively, theprocess may be performed for a plurality of routes and the selected configuration is one for which the SFOC is lowest for a route that is based on the emissions data (e.g. for a route for which the associated emissions are determined to be below a predetermined threshold). Therefore, in some examples the process 300 may be performed for a route, or for a plurality of routes, that meet an emission restriction (e.g. whose total emissions are less than a predetermined emissions threshold).

[0111] Any obtained emissions data may be utilised in combination with data from the energy planner as discussed above as follows.

[0112] From the energy planner (which comprises data describing at least one route segment the vessel may traverse) it may be determined the vessel current position and the remaining distance to a vessel destination (which may be associated with the vessel objective). The remaining distance may include at least one of: a planned speed or heading between waypoints, an estimated energy demand between the waypoints, or any waypoints that cross an emissions control area (e.g. an NOx or SOx emission control area) or an CH4 emission restriction area.

[0113] Effectively the process can then output a vessel power configuration that meets an emissions restriction or target and for which the SFOC is the lowest.

[0114] Of course, the SFOC may be linked to emissions and in some examples the outputted SFOC may be associated with an emission level (e.g. NOx, SOx, and / or a CH4 emission level) and the selected configuration may be one for which the SFOC and / or its associated emission level is the lowest. Thereby, optimum power system configurations that meet emissions targets may be determined.

[0115] Actual vessel operation, it will be appreciated, may comprise any one or more of: running the vessel at reduced speed to minimise engine load and emissions, utilizing batteries if available to support for the reduced engine power while still maintaining the speed and stay within emission, and providing information to ship power management system for adjusting the battery parameters like SoC setpoint, discharge limits etc.

[0116] Any emissions level or data may be one, or any combination, of SOx, NOx, and CH4. In general, the process 300 may be performed for a specific route, e.g. a route may be an input into the process and the process may be performed forthat route, or the process 300 may be performed fora plurality of routes and the process may be performed forthose plurality of routes. In the former example one SFOC may be generated and in the latter example a plurality of SFOCs may be generated (the lowest SFOC-route being selected at 306). This allows the process to check a SFOC for one or more routes and then a specific route (the one having the lowest SFOC) may be selected (at 306).

[0117] The SFOC may comprise a measure of fuel oil consumed by the vessel while operating in a machine mode according to a particular power output configuration (optionally at a targetspeed) to achieve a vessel objective. The SFOC may be determined by summing the individual SFOCs for each power generator that is on and used in a given machine mode, in other words the SFOC is a total SFOC for the power output configuration and power system. The SFOC may be determined based on the power of the or each power generator, the power rate of fuel consumption and the amount of energy produced. The SFOC may comprise a measure of fuel usage. The SFOC may be determined by dividing the rate of fuel consumption by the amount of energy produced. In one example the SFOC may be a quantity obtainable by, or obtained by, or equal to, dividing the fuel consumption rate by the amount of power produced and multiplied by the fuel density. In one example the SFOC may be according to the following equation:

[0118] SFOC = (F*p) / P

[0119] where F is the fuel flow, p is the fuel density, and P is the amount of power produced. In an example the fuel flow F may be given in litres per hour (L / h), the density p may be given by grams per litre (g / L) and the power may be given in kilowatts (kW). In this example the SFOC may be given in grams per kilowatt-hour (g / kWh). Alternatively, the SFOC may be provided for each individual power generator, for example, by the manufacturer, and may be used to estimate the fuel consumption of each power generator. In this way, the total SFOC for the vessel can be estimated without a fuel flow for individual power generator. The SFOC may therefore be based on a stored value (e.g. based on a manufacturer’s estimate) and the stored value may comprise a SFOC table. The table may be updated continuously or periodically or may be updated when a determined SFOC deviated from the stored SFOC deviates by more than a predetermined threshold which may trigger a request for a new SFOC value or alert a user to provide one. On reason the stored SFOC may deviate from the stored SFOC is due to engine utilisation or wear and tear. The SFOC determined at 304 may therefore be based on a stored value such a manufacturer’s estimate, and this may be stored in a look-up table against differing values of power demand data, and the process may optionally comprise updating the stored value, e.g. when it a current value differs too significantly from the stored value.

[0120] As stated above with reference to Figure 2, a machine mode of the vessel corresponds to a type of powertrain to deliver power to the vessel for the vessel to achieve the vessel objective. One example of a machine mode is a diesel mode. In the diesel mode a diesel powertrain may be used to generate power (e.g. a powertrain driven by diesel fuel). Another example of a machine mode is a hybrid mode. In the hybrid mode mechanical power from the engine is used in combination with one or more battery units, ora switchboard, to use electrical energy to drive the propulsors. Further examples of machine modes are a diesel-electric mode and a diesel-mechanic mode. In a diesel-electric mode mechanical force of the diesel engine may be converted into electrical energy (e.g. via an alternator), and using the electrical energyto drive the propulsors. In a diesel-mechanic mode mechanical force produced by the diesel engine may be used (e.g. directly) to drive the propulsors. This list is not exhaustive and other machine modes may be apparent to the skilled person. For example, in a hybrid mode one propeller may be driven by an engine and the other propeller may be driven by the switchboard. A hybrid mode may therefore describe a mode in which mechanical power is used in combination with electrical power to drive one or more propulsors and the electrical power may be provided by one or more batteries and / or one or more other sources (e.g. a switchboard).

[0121] It will therefore be appreciated that a given machine mode corresponds to a number of power system components that are on or active and working to form a powertrain to deliver a type of power to at least one propulsor and in each machine mode the vessel power system can operate in a plurality of power output configurations, each power output configuration corresponding to respective power output levels of at least one generator that is active in the machine mode. For example, a diesel mechanic mode may correspond to the two main engines being on, or active, or used etc. and each power output configuration within this machine mode may correspond to respective power output levels of each of the main engines. A diesel mechanic mode may correspond to the propeller being driven mechanically through the shaft connected to an engine (e.g. any engine such as one or both main engines may not be ON, an auxiliary engine may be used instead or in addition). A diesel mechanic mode may therefore correspond to a propeller being driven by mechanical energy. In this way a diesel machine mode may correspond to a mechanical powertrain and a hybrid machine mode may correspond to an electrical powertrain being used in combination with a mechanical powertrain (e.g. a hybrid powertrain).

[0122] In one example, if there were two main engines then a first power output configuration may correspond to main engine 1 (“ME1”) operating to deliver a 30% of the total power demand and main engine 2 (“ME2”) operating to deliver 70% of the total power demand; and a second power output configuration may correspond to ME1 operating to deliver 40% of the total power demand and ME2 operating to deliver 60% of the total power demand etc. In another example a power output configuration may correspond to one ME driving the propeller at a first power output level and delivering power to the switchboard via the shaft while another ME is driving the other propeller at a second power output level. In yet another example a power output configuration may correspond to ME delivering power to one of the propellers at a first mechanical power output level and delivering power to the switchboard via shaft with other propeller is running in “PTI mode” where it is driven by a motor or PTI supplied by electrical power at a first electrical power output level from an auxiliary engine. Examples of other power output configurations would be understood by the skilled person, each power output configuration being operable to produce a total power (e.g. the sum of the poweroutputs of each operating power generator) that is equal to the power demand as described by the power demand data.

[0123] As indicated by the dotted block 306, the process may comprise selecting the power output configuration that is associated with the lowest SFOC. This effectively determines the machine mode in which the vessel power system will operate to powerthe vessel. As indicated by the dotted block 308, the process may comprise transmitting the power output configuration (and therefore the machine) that is associated with the lowest specific fuel oil consumption. The process may be performed remotely from the vessel, in which example block 308 may indicate that the system or circuitry etc. executing the process transmits the selected power output configuration having the lowest SFOC to the vessel for the vessel to implement the power system configuration (e.g. the machine mode and power output configuration) corresponding to the lowest SFOC. Alternatively, the process may be performed on-board the vessel by a system that is not configured to control the power system, in which case the process transmits the selected power system configuration to a system responsible for controlling the operation of the power system (e.g. an electronic control unit).

[0124] As indicated by the dotted block 310, the process may comprise causing the power system to be operated in the machine mode and according to the power output configuration associated with the lower specific fuel oil consumption (e.g. in examples where the process is performed by an on-board controller that is also responsible for controlling the power system).

[0125] As indicated by the dotted block 312 the process may comprise obtaining vessel objective data indicating the objective that the vessel is to achieve. As above, this data may comprise data relating to a transit operation of the vessel. The objective may alternatively or additionally comprise a measure of operator preference e.g. at least one of a reduction in fuel consumption, an improvement in transit time, and an improvement in vessel motions, for example. Alternatively or additionally the vessel objective may comprise at least one of determining that the vessel is not in “dynamic positioning” (“DP”) mode and determining that the vessel is not in “anchor handling” (“AH”) mode. Alternatively or additionally the vessel objective may comprise determining that the vessel speed is above a threshold (e.g.4 knots).

[0126] As indicated by the dotted block 314 the process may comprise obtaining operational data indicating a set of candidate operational modes (synonymous with “machine modes”) for the vessel, the plurality of machine modes for which the SFOC is determined being the set of candidate operational modes. This block may comprise receiving the machine modes (e.g. where these are known) or determining the machine modes that a given vessel can adopt (e.g. this may not be known ahead of time). Block 314 of the process may be utilised in particular when the process is being implemented off-board a vessel, e.g. by a remote controller.As indicated by the dotted block 316 the process may comprise obtaining vessel operational data, wherein the specific oil consumption is determined for each machine mode and power output configuration based on the vessel operational data. Vessel operational data may comprise information relating to each component of the power system and / or fuel flow and / or any information that is specific to the operation of the vessel.

[0127] As indicated by the dotted block 318 the process may comprise obtaining data indicating the current speed of the vessel and block 304 may accordingly comprise determining, for each machine mode and power output configuration, the SFOC based on a current vessel speed. The current vessel speed may be given in knots or as a speed through water (“STW”) or as any other appropriate measure such as speed over ground (“SOG”) which may be determined via GPS data and which may provide an indication as to the ocean current.

[0128] As indicated by block 320, and as discussed above, in some examples the objective that the vessel is to achieve is associated with the vessel operating in transit. In some examples the process 300 is performed while the vessel is operating in the transit mode. In some examples, the process comprises obtaining input indicating that the vessel is to be operating, or is currently operating, in the transit mode.

[0129] As indicated by the bracketed text next to the looping arrow, in some examples it may be determined at block 304, for each machine mode and power output configuration, the SFOC for the vessel to achieve the objective and to meet the power demand for a plurality of different vessel speeds. In this example, a SFOC may be determined not only for a plurality of machine modes and power output configurations per machine mode, but also for a plurality of different speeds at which the vessel could travel. In this way, whilst the process 300 could be performed for a given vessel speed (e.g. received as input - for instance it may be determined that the vessel is to travel at a specific transit speed and then the process is performed to determine the SFOCs for various power output configurations and machine modes for that transit speed), the process 300 could also be performed to determine which vessel speed, power output configuration and machine mode that is associated with the lowest SFOC (e.g. various transit speeds may be acceptable at which the vessel could travel). The former example may be used where a target vessel speed is desired, latter example may be utilised where a range of target vessel speeds are possible.

[0130] As mentioned above, the process may comprise obtaining vessel operational data. Such operational data may comprise data relating to the machine modes (e.g. the type of powertrain) of the vessel. By way of example, one given vessel may not have the machinery to operate as a hybrid system whereas another given vessel may have that functionality. Operational data may contain this information. The process may comprise determining the operational data, which may effectively comprise determine which operational modes (e.g. which machine modes, e.g. which types of energy) are possible for a given vessel.It will be appreciated that the process 300, indeed all processes described herein, may be implemented in control management software for the vessel, such as control software for an energy management system of the vessel and / or a power system management system of the vessel, and this may be implemented on-board the vessel or remotely. Implementing the process will result in the vessel’s power system being controlled according to a selected machine mode and engine configuration. Any or all of the blocks of the algorithm of the process 300 (for example 304 and / or 306) may be running in a server layer and not in the real-time control layer of a computing system configured to implement the process. Some or all of the blocks of the process 300, indeed all processes describer herein, may also be run in real-time. As stated above, any or all parts of the processes described herein may be performed by a controller.

[0131] Figure 4 shows a process 400 that illustrates part of the process 300 (although for the avoidance of doubt the process 400 may be implemented either as a separate process to 300 or as part of the process 300). The process 400 may be to determine a SFOC.

[0132] At 401 the process comprises obtaining at least one input. The input may comprise any suitable input to enable a set of candidate machine modes to be determined at the forthcoming step of the process. For instance, powertrain and / or power system and / or power system component information may be received as input at block 401. Additionally, vessel speed information (e.g. transit speed) or any other vessel parameters may be received at block 301. Any input necessary to determine machine modes and power output configurations may also be received as an input at block 401 , such as power demand data (e.g. route-data and / or weather conditions and / or non-propulsion loads) and / or emissions data and / or SFOC data (e.g. from a manufacturer). Power demand data may be received from an energy planner as discussed above.

[0133] At 402 the process comprises determining the candidate machine modes for at least one given vessel. This may be done based on the one or more inputs obtained at 401.

[0134] Block 403 represents the output of block 402 of the process 400. Here, it is shown that n candidate modes are possible for the given vessel with respect to which the process is performed.

[0135] At 404 the process comprises determining the candidate power output configurations for each mode determined at 402.

[0136] Block 405 represents the output of block 404 of the process 400. Here, it is shown that for the first machine mode MM1 , r power output configurations are possible, for the second machine mode MM2, s power output configurations are possible, and for the nth machine mode MMn, t power output configurations are possible.

[0137] At 406 the process comprises determining the SFOC for each power output configuration in block 405.The process 400 may be at least part of block 304 of the process 300. Accordingly in some examples after block 406 the process may turn to block 306 of the process 300 where a power output configuration and a machine mode is selected (for example, the power output configuration and machine mode corresponding to the lowest SFOC). Some examples of the process 300 may then comprise selecting the vessel speed, power output configuration and machine mode associated with the lowest SFOC.

[0138] For example, the selected engine configuration and machine mode may be Cpq - e.g. the qth power output configuration of the pth machine mode, this being the one with the lowest SFOC. If there are n engines in a given machine mode then there may be 2n-1 power output configurations.

[0139] It will further be appreciated that in examples where the SFOC is to be determined for power output configurations and machine modes at different speeds, then there will be further blocks of the process, e.g. there will be a first power output configuration for the first machine mode at a first speed, and a first power output configuration for the first machine mode at a second speed etc...

[0140] The processes 300 and 400 may receive data, e.g. various inputs, from any management system on board the vessel to determine the SFOC. These inputs may be vessel speed, the power of each power producer that is part of the vessel’s power system, and the fuel flow. The data may be processed according to the machine modes that the vessel is capable of operating (e.g. whether the vessel can operate in a hybrid mode etc.).

[0141] Determining the SFOC (e.g. at block 304 or 406) may be performed, in some examples, via a ML model trained to output an SFOC. Examples using a ML model to determine the SFOC will now be described.

[0142] Figure 5 schematically illustrates a ML model generally, at 500. Block 500 may be used as part of the process 300 or 400, e.g. blocks 304 or 406 may comprise block 500. As Figure 5 shows, the ML model 500 receives a power demand as input. The power demand may be a power demand on the vessel’s power system in order to power the vessel to achieve and / or maintain a target speed. Optionally, a vessel speed, (as used herein vessel “speed” should be understood as any suitable measure of speed for the vessel, such as a STW or a SOG) may also be input to the model (speed data may be part of the power demand data or separate data). The model 500 may be trained to output the SFOC for a given machine mode and / or for a given engine configuration.

[0143] Figure 5 and block 500 is intended to represent a schematic general model, more specific examples will be given with respect to Figures 6 and 7.

[0144] Block 500 may receive a power output configuration as input and may be trained to output a SFOC (based on receiving the power demand as input, and optionally speed data) for that power output configuration. Alternatively or additionally, block 500 may receive amachine mode as input and may be trained to output a SFOC (based on receiving the power demand as input, and optionally speed data) for a given machine mode. Therefore, block 500 may receive a power output configuration for a machine mode as input and may be trained to output a SFOC (based on receiving the power demand as input, and optionally speed data) for a plurality of power output configurations for the given machine mode operating according to the power output configuration. The model may also receive any constraints and / or preferences as input. The model may receive power demand data (e.g. route-data and / or weather conditions and / or non-propulsion loads) and / or emissions data and / or SFOC data (e.g. from a manufacturer) as input. Power demand data may be received from an energy planner as discussed above.

[0145] Therefore, in some examples, a single ML model 500 may be used, in which case the model may be trained based on historic data relating to SFOCs for different power output configurations of different machine modes, the current power demand of the marine vessel, optionally the vessel speed, the ML model being configured to output a SFOC for a plurality of power output configurations and a plurality of machine modes based on receiving the current power demand as input, and optionally vessel speed. In this example, the ML model is able to receive the power demand as input (and optionally vessel speed) and determine the SFOC for a plurality of power output configurations for a plurality of machine modes (e.g. the output indicated by block 405 of Figure 4). However, in another example, an ML model is configured to output a SFOC for a plurality of power output configurations for a given machine mode based on receiving the current power demand as input, and optionally the vessel speed. In this example, block 500 may represent a plurality of ML models, one for each machine mode, where each mode is able to receive the power demand as input, and optionally vessel speed, and determine the SFOC fora plurality of power output configurations for its respective machine modes (e.g. one ML model per column of the output indicated by block 405 of Figure 4, so block 500 represents n machine modes in the Figure 3 example). Where the ML model is specific to a machine mode, in addition or in alternative to receiving the power demand and vessel speed as input, the ML model may instead receive a plurality of output configurations as input (e.g. from a “splitter”), the ML model being configured to determine a SFOC for each power output configuration that it receives. This will be discussed with reference to Figures 6 and 7 that use a “splitter.”

[0146] Figure 6 schematically shows part of a process that comprise a machine learning (“ML”) model (or ML algorithm), the ML model being indicated at 600, that uses the ML model to determine the SFOC.

[0147] Block 601 is a splitter. The splitter is configured to determine, based on receiving a power demand as input, a plurality of power output configurations (e.g. respective power output levels of power generators of the vessel power system) such that the power system willdeliver the total power demand. For example, the splitter is configured to determine the percentages of the total power demand that will be delivered by each generator. For example, for a given power demand, the output of the splitter may be that ME1 may be operated to deliver 40% of the total power demand and ME2 may be operated to deliver 60%, that ME1 may be operated to deliver 30%, and ME2 at 70% etc. The output of the block 601 may be the “C” columns indicated in block 305 of Figure 4, e.g. Cpq forgiven values of p and q. The splitter 601 is therefore configured to “split” the loads between the power generators to achieve the total power demand, or is configured to distribute the power demand between the power generators to determine the possible power output configurations of the vessel system operating in a given machine mode that can achieve the total power demand. Power demand data (e.g. route-data and / or weather conditions and / or non-propulsion loads such as hotel loads) and / or emissions data and / or SFOC data (e.g. from a manufacturer) may be received as inputs to the model 600 with the power demand data may be received from an energy planner as discussed above.

[0148] The output of the splitter 601 is the input to the ML model 600. In other words, the ML model 600 is configured to receive at least one power output configuration (e.g. a plurality of power output configurations) as input. The ML model 600 is configured to determine a SFOC for each power output configuration configuration received as input. As Figure 6 shows, the output of the splitter may be a percentage of the total power demand that is to be delivered by each of a plurality of power generators (which may also be referred to as power producers). As also indicated by Figure 6, the ML model 600 may optionally receive the configurations of a number (e.g. each) of the power producers as input, this may comprise a number of available power generators and / or their operational status (e.g. operational, unbroken, a maximum capacity, rated performance etc.). The ML model 600 may receive a vessel speed as input (which may be a transit speed and / or a minimum vessel speed to be achieved and / or maintained).

[0149] The ML model 600 is trained to output, based on receiving the above-described parameters as inputs, an SFOC for each power output configuration received as input.

[0150] The process, at 603, comprises selecting the power output configuration that has the minimum SFOC.

[0151] The process, at 604, comprises outputting that power output configuration having the minimum SFOC, and the vessel may be controlled according to this selected power output configuration in some examples.

[0152] Block 600 may be configured to (trained to) output an SFOC for a plurality of power output configurations, based on receiving the power output configurations as input, for a given machine mode. Alternatively, block 600 may be configured to (trained to) output an SFOC for a plurality of engine configurations and for a plurality of machine modes, based on receivingthe power output configurations and machine modes as input. Therefore, block 600, may be configured to determine SFOCs fora plurality of machine modes orfor a given machine mode. The selected power output configuration may be transmitted, e.g. to another entity, to enable the vessel power system to be controlled according to the selected power output configuration in some examples.

[0153] Figure 7 also schematically shows part of a process that comprise a machine learning (“ML”) model (or ML algorithm), the ML model being indicated at 700, that uses the ML model to determine the SFOC.

[0154] The Figure 7 process uses a plurality of ML models, one for each machine mode. Five are shown in this example but this is for illustrative and example purposes only and is therefore not exhaustive. Each ML model is trained to output the SFOC for a plurality of power output configurations (received as input) for a given machine mode. The five are: a ML model 700-1 trained to output the SFOC for a plurality of power output configurations in the DMNormal Mode, a ML model 700-2 trained to output the SFOC for a plurality of power output configurations in the DM Speed Mode, a ML model 700-3 trained to output the SFOC for a plurality of power output configurations in the DM Economy Mode, a ML model 700-4 trained to output the SFOC for a plurality of power output configurations in the DE mode, and a ML model 700-5 trained to output the SFOC for a plurality of power output configurations in the Hybrid Mode. Power demand data (e.g. route-data and / or weather conditions and / or nonpropulsion loads such as hotel loads) and / or emissions data and / or SFOC data (e.g. from a manufacturer) may be received as inputs to any of the models 700-1 , ..., 700-5 with the power demand data may be received from an energy planner as discussed above.

[0155] According to the Figure 7 process, each splitter for each machine mode that is part of the process receives power demand data as input. Optionally, each ML model 700-1 , ..., 700-5 receives, in addition to the output of the splitter, vessel speed data and / or power producer configuration data, e.g. as described with reference to Figure 6 (not shown in Figure 7 for brevity and ease of illustration). As for the splitter 601 described with reference to Figure 6, each splitter 701-1 , ..., 701-5, is configured to output a plurality of power output configurations so that the power demand can be achieved. This is schematically indicated in Figure 7. So, for instance, the output of 701-1 are the possible operations of ME1 and ME2 (e.g. what percentages of the demand these can be operated to deliver, e.g. ME1 delivering 70% and ME2 delivering 30% in a first engine configuration, ME1 delivering, 40%, ME2 delivering 60% in a second engine configuration etc.).

[0156] At 703 the process comprises determining the power output configuration associated with the minimum SFOC for that machine mode. At 703’ the process comprises determining the power output configuration of those associated with the minimum SFOC. It will therefore be appreciated that the output of block 703’ will be the power output configuration and machinemode associated with the minimum SFOC. Block 704 represents the output of the process and may comprise operating the vessel according to the power output configuration and the machine mode associated with the minimum SFOC. The selected power output configuration may be transmitted, e.g. to another entity, to enable the vessel power system to be controlled according to the selected power output configuration in some examples.

[0157] As stated above, although not shown in Figure 7, each ML model 700 may be trained to output the SFOCs additionally based on receiving power producer configuration data and / or vessel speed data as input.

[0158] It will be appreciated that any of the models described herein may be stored, trained, and ready for use when the SFOC is to be determined. In examples, the machine learning models are trained based on historic data relating to specific fuel oil consumptions for different engine configurations for the vessel operating during transit, which will now be described. Such training data may comprise power demand data (e.g. route-data and / or weather conditions and / or non-propulsion loads such as hotel loads) and / or emissions data and / or SFOC data (e.g. from a manufacturer) with the power demand data may be received from an energy planner as discussed above.

[0159] Deriving a ML Model

[0160] Figure 8 shows process 800 that may be performed by a controller (such as 2 or 3 as described above) and may be performed in conjunction with any of the processes described above. The process 800 may be a process for obtaining training data to train a ML model, such as any one of the ML models discussed above.

[0161] At 802 the process comprises obtaining vessel operational data, for example data that indicates at least one of: the vessel speed, an operational state of at least one component of the vessel power system (e.g. the generator), fuel flow, SFOC, a power output configuration of the vessel system, and a machine mode. At 804 the process comprises determining a machine mode in which the vessel is operating. At 806 the process comprises determining a number of generators of the vessel that are running and at what power output configuration. At 808 the process comprises determining the specific fuel oil consumption of the vessel for the power output configuration and machine mode. At 810 the process comprises storing the specific fuel oil consumption, the power output configuration, and the machine mode, as training data. Therefore, Figure 8 indicates that if, from a given data set, the SFOC is derivable for a given vessel operating in a given power output configuration in a given machine mode to achieve a given power demand (optionally for a given vessel speed), then this data may be suitable as training data to train a ML model. Given that the ML is to be trained to output a SFOC based on receiving at least a power output configuration as input, suitable training datamay comprise, as a minimum, an SFOC for a particular power output configuration to achieve a given power demand value. Optionally, the data may comprise vessel speed, power producer (e.g. generator) configurations, and a machine mode for the vessel. In some examples the data selected for training data may relate to the vessel being operated in its transit phase. Figure 800 may be a process of storing data collected ‘on-the-fly’ as data suitable to be used to train a machine learning model at a later time.

[0162] The process 800 may be considered part of a “data preparation” step where steps 804-808 comprise filtering data.

[0163] Figure 9 shows a schematic process for deriving a ML model.

[0164] At 901 the process comprises obtaining data and at 902 the process comprises a data preparation step. The data preparation step 902 may comprise part of the Figure 8 process. The output of the data preparation step is suitable training data sets, e.g. suitable data to train a ML model. The training data may comprise data sets each comprising a SFOC for a given power output configuration of the vessel power system to achieve a power demand. The training data may comprise data sets each comprising a SFOC for a given power output configuration and machine mode to achieve a power demand. The training data may comprise data sets each comprising a SFOC for a given power output configuration and machine mode and vessel speed to achieve a power demand.

[0165] Block 903 schematically indicates the process of deriving a predictive ML model comprising training the model using the training data, 904, evaluating the mode, 905, and fine-tuning the model, 906. The output of block 903 is the predictive ML model 908 that may be tested in a process where it receives test data 907 to produce an output 909. The predictive ML model 908 may be used as any of the ML models discussed hereinabove. It will therefore be appreciated that the ML model 908 may be trained on historic data (comprising data relating to at least an SFOC for particular power output configurations to achieve a given power demand in a given machine mode and for a given vessel speed) to output a SFOC based on receiving a given power demand and engine configuration as input.

[0166] Figure 10 shows an example tangible and non-transitory machine-readable medium 1000 in association with a processor 1002. The medium 1000 comprises a set of instructions 1001 that, when executed by the processor 1002, cause the processor 1002 to perform any of the processes described herein (e.g. any one of the blocks thereof). The machine-readable medium may be a computer-readable medium. The instructions may comprise program code. The processor may be any hardware suitable to perform the processes described herein, such as a central processing unit (CPU) or microprocessor. The term “processor” should be interpreted broadly to include a CPU, a controller or microcontroller, processing circuitry, digital signal processor, field programmable gate array (FPGA), and / or an application specific integrated circuit (ASIC). The term “processor” should also be interpreted as including multipleprocessors and, as such, the processes described herein may be distributed to and performed by several processors. The machine-readable medium may be a separate physical entity to the processor. Alternatively, the machine-readable medium may be part of the same system as the processor. The instructions may be stored in a memory and to execute the method the processor may access the memory. The machine-readable medium and / orthe processor may be located on-board a marine vessel or remotely (for example as part of a server computing system that is part of a cloud computing environment) depending on the implementation. The machine-readable medium may be any suitable medium for storing executable instructions, such as but not limited to an electronic storage device, for example a Random Access Memory (RAM), a Read-Only Memory (ROM), EPROM, EEPROMs, flash memory, CD-ROM, HDD, SDD, a storage device or optical disk having computer-readable program code thereon.

[0167] Figure 11 shows a system 1100 for performing the processes described herein. Specifically, the system comprises a vessel power system configuration determination module 1101 that is configured to perform the processes described herein (e.g. any of the blocks thereof), in particular the module 1101 is configured to determine a configuration for a vessel power system according to the processes 200-700 with reference to Figures 2-7. The system 1100 is shown in conjunction with a vessel power system control module 1102 and the vessel power system 1104 (e.g. 200 with reference to Figure 2). Although shown separately, the vessel power system control module 1102 may be part of the same system 1100 as the vessel power system configuration determination module 1101. As shown by the dotted lines, the configuration determination module 1101 is configured to receive data from the vessel power system 1200 (e.g. power producer data describing a property of a power generator of the vessel power system 1104). The configuration determination module 1101 is further configured to determine the power output configuration of the power system 1104 (and therefore the machine mode) associated with the lowest SFOC and configured to transmit this configuration to the vessel power system control module 1102. Upon receipt, the vessel power system control module 1102 is configured to operate the vessel power system 1104, e.g. cause the power system to operate, in the machine mode and according to the power output configuration that is associated with the lowest SFOC.

[0168] The vessel power system 1104 is onboard the marine vessel. However, the system 1100 (comprising the configuration determination module 1101) may be onboard or offboard the vessel. Indeed, the configuration determination model 1101 may be remote from the marine vessel and may transmit the determined lowest SFOC configuration to the power system control module 1102 to control the vessel. The control module 1102 may be on-board the vessel in some examples although in other examples the control module 1102 may be remote and may remotely control the vessel power system 1104.Figure 12 shows schematically an example cloud computing environment 1200 that may comprise the system 1100 of Figure 11. The environment 1200 comprises a cloud network 1202 that may comprise one or more server systems, or cloud service provider, or cloud computing infrastructure, or cloud service provider, denoted schematically by 1203, and a number of client devices 1201-1 , 1201-4 that are connected (e.g. interconnected) to one another through the cloud network, for example via the internet. Via the cloud computing environment 1200, software may be provided to the devices 1201 on-demand, for example the cloud environment can deliver the processes described herein as executable instructions (e.g. as “software as a service” (SaaS)) over the internet. The system 1100 as described above with reference to Figure 11 may be part of any of the client devices 1201 or may be part of the server 1203. For instance, 1201 may be a marine vessel, or 1201 may denote a controller for a marine vessel (being on-board or remote). Either may comprise the system 1100, e.g. the modules 1101 , 1102, and / or 1104. Alternatively or additionally the server 1203 may comprise the system 1100 (e.g. the module 1101 and / or 1102), the server being configured to determine a power output configuration and machine mode associated with the lowest SFOC for a given vessel and transmit that configuration to the vessel for the vessel to adopt in these examples. It will be appreciated that the server 1203 may be associated with a number of vessels, and may be configured to determine a power output configuration and machine mode associated with the lowest SFOC for a plurality of vessels and transmit the respective power output configuration associated with the lowest SFOC to each vessel. The machine-readable medium as discussed above with reference to Figure 10 may be part of any of the client devices 1201 (e.g. a vessel or on-board or remote controller etc.) or may be part of the server device 1203.

[0169] As used herein, “obtaining” data as used herein may comprise receiving data (e.g. from a sensor), measuring data (e.g. by a sensor), determining or calculating or estimating data (e.g. from data received from a sensor), or retrieving data (e.g. from a memory storing data), depending on the example.

[0170] The processes, systems, and instructions described herein may be implemented according to any suitable hardware and / or software combination sufficient to cause the processes described herein to be executed. For instance the systems, and instructions may be implemented on, or on any suitable combination of, a digital signal processor, field programmable gate array, and / or application specific integrated circuit (ASIC). Any one or more of the blocks of the processes described herein may be performed by processor or processing circuitry as discussed above, the processor may be configured to execute instructions, such as processor control code, that, cause a controller to operate according to the processes described herein. Such instructions may be stored on a non-transitory machine-readable medium. Such instructions may be stored in a memory. Such instructions may bestored on any suitable memory medium, e.g. on a volatile or non-volatile medium, programmed memory (e.g. read-only memory such as firmware), or a data carrier. The processing circuitry may comprise such a memory storing the instructions. In examples where a compressor performance map is used to determine the compressor’s proximity to the surge state, this may also be stored in the same memory as the operating instructions. In other words, a non-transitory machine-readable medium may store instructions that, when executed by processing circuitry, cause the processes herein to be performed. The instructions may comprise code or microcode. The instructions, when executed, may be in any suitable programming language to allow the controller to be dynamically configured and / or reconfigured. The controller and / or processing circuitry may equally comprise, and may therefore be referred to as, a processor, microcontroller or microprocessor.

[0171] The person skilled in the art realizes that the present disclosure by no means is limited to what is explicitly described above. On the contrary, many modifications and variations are possible within the scope of the appended claims. Additionally, variations can be understood and effected by the skilled person in practicing the claimed invention, from a study of the drawings, the disclosure, and the appended claims.

Claims

CLAIMS1. A computer-implemented method for determining a configuration for a vessel power system, the vessel power system comprising a plurality of power system components including:a plurality of power generators, each power generator configured to generate a type of power, andat least one propulsorto cause the vessel to move, each propulsor configured to be powered by power from at least one power generator,the power system being operable to cause a type of power to be delivered from a power generator to a propulsorto drive the propulsor, the power system being operable in a plurality of machine modes, each machine mode corresponding to respective configurations of the power system components such that a type of power is delivered from at least one power generator to at least one propulsor,the method comprising:obtaining power demand data indicating a current power demand of a marine vessel to achieve a vessel objective; anddetermining, for each one of a plurality of machine modes and for a plurality of power output configurations for each machine mode, a specific fuel oil consumption for the vessel to achieve the objective and for the vessel power system to meet the power demand, wherein each power output configuration for each machine mode corresponds to respective power output levels of each power generator configured to deliver power to a propulsor in that machine mode, the sum of the power output levels for each power generator in each power output configuration being equal to the current power demand.

2. The computer-implemented method of claim 1 , the method further comprising:transmitting the machine mode and power output configuration that is associated with the lowest specific fuel oil consumption to a vessel power system control unit.

3. The computer-implemented method of claim 1 or 2, the method further comprising:causing the power system to be operated in the machine mode and according to the power output configuration associated with the lower specific fuel oil consumption.

4. The computer-implemented method of any preceding claim, the method further comprising:obtaining vessel objective data indicating the objective that the vessel is to achieve.

5. The computer-implemented method of any preceding claim, the method further comprising:obtain operational data indicating the machine modes in which the power system of the vessel can be configured to operate.

6. The computer-implemented method of any preceding claim, wherein the objective that the vessel is to achieve is associated with the vessel operating in a transit mode.

7. The computer-implemented method of any preceding claim, the method further comprises:obtaining input indicating that the vessel is to be operating, or is operating, in a transit mode.

8. The computer-implemented method of any preceding claim wherein determining the specific fuel oil consumption, for each one of a plurality of machine modes and for a plurality of power output configurations for each machine mode, is based on a speed of the vessel.

9. The computer-implemented method of claim 8, the method further comprising:obtaining data indicating the speed of the vessel.

10. The computer-implemented method of any of claims 1-9, the method further comprising:determining, for each machine mode and for each power output configuration, the specific fuel oil consumption for the vessel to achieve the objective and to meet the power demand for a plurality of different vessel speeds.

11. The computer-implemented method of claim 10, the method further comprising: transmitting the machine mode and power output configuration and vessel speed that is associated with the lowest specific fuel oil consumption to a vessel power system control unit.

12. The computer-implemented method of claim 10 or 11, the method further comprising:causing the power system to be operated in the machine mode and according to the power output configuration, to thereby cause the vessel to be operated at the associated vessel speed, that is associated with the lowest specific fuel oil consumption.

13. The computer-implemented method of any preceding claim, the method further comprising:determining, for each machine mode and for each power output configuration, the specific fuel oil consumption by:causing the current power demand to be input into a machine learning model trained based on historic data relating to specific fuel oil consumptions for a plurality of power output configurations and for a plurality of machine modes to output at least one specific fuel oil consumption associated with operating the vessel power system in at least one machine mode according to at least one power output configuration to meet the current power demand of the vessel based on receiving the current power demand of the vessel as input; andobtaining the specific fuel oil consumption output by the machine learning model.

14. The computer-implemented method of any preceding claim, the method further comprising:determining, for each machine mode and for each power output configuration, the specific fuel oil consumption by: causing the current power demand of the vessel to be input into a plurality of machine learning models, each machine learning model in the plurality being trained based on historic data relating to specific fuel oil consumptions for a plurality of power output configurations in a respective machine mode to output a plurality of specific fuel oil consumptions associated with operating the vessel power system in the respective machine mode according to the plurality of power output configurations to meet the current power demand of the vessel based on receiving the current power demand of the vessel as input; andobtaining the specific fuel oil consumption output of each machine learning model for each machine mode.

15. The computer-implemented method of claim 13 or 14, wherein the or each machine learning model is trained based on historic data relating to specific fuel oil consumptions for a plurality of power output configurations and for the or each machine mode for the vessel operating in a transit mode.

16. The computer-implemented method of any preceding claim wherein the plurality of machine modes comprises at least one of:a diesel mechanic mode in which at least one power generator is configured to generate mechanical power to power at least one propulsor,a diesel electric mode in which at least one power generator is configured to generate electrical power to power at least one propulsor, anda hybrid mode in which at least one power generator is configured to generate mechanical power to power at least one propulsor at least one power generator is configured to generate electrical power to power at least one propulsor.

17. The computer-implemented method of any preceding claim, wherein the power generators of the vessel power system comprise one or more of: a main engine configured to generate mechanical power, an auxiliary engine configured to generate electrical power, a generator configured to generate electrical power, and a motor configured to generate mechanical power.

18. The computer-implemented method of any preceding claim, the method further comprising:obtaining vessel operational data, wherein the specific fuel oil consumption is determined for each machine mode and for each power output configuration based on the vessel operational data.

19. The computer-implemented method of claim 18, wherein the vessel operational data comprises at least one of: information relating to at least one component of the vessel power system and the flow rate of fuel to at least one power generator.

20. The computer-implemented method of any preceding claim, the method further comprising:obtaining data indicating at least one of: the vessel speed, an operational state of at least one component of the vessel power system, and the flow rate of fuel flow to at least one power system;determining a machine mode in which the vessel is operating;determining a number of power generators of the vessel power system that are running according to respective power output levels;determining the specific fuel oil consumption of the vessel; andstoring the specific fuel oil consumption, the power output configuration being the respective power output levels of the power generators, a route or a plurality of routes the vessel could traverse, a weather condition or set of weather conditions the vessel could be subject to while traversing a given route, power demand data as described above, an emission (e.g. greenhouse-gas or carbon) emission for the vessel while traversing a given route, and the machine mode, as training data.

21. The computer-implemented method of any preceding claim wherein the power demand data comprises a power demand for a vessel to traverse a given route in a given weather condition, the weather condition being based on weather data.

22. The computer-implemented method of any preceding claim wherein the power demand data comprises a total non-propulsion load data comprising an determined power required to operate at least one non-propulsion load of the vessel along a given route.

23. The computer-implemented of any preceding claim wherein the determined SFOC is based on an emissions value for a given route the vessel is to traverse, the determined SFOC being based on a route associated with an emissions value that is less than a predetermined threshold.

24. A machine-readable medium storing instructions that, when executed by processor, cause the processor to perform the method of any preceding claim.

25. A system for determining a configuration for a vessel power system, the vessel power system comprising a plurality of power system components including:a plurality of power generators, each power generator configured to generate a type of power, andat least one propulsorto cause the vessel to move, each propulsor configured to be powered by power from at least one power generator,the power system being operable to cause a type of power to be delivered from a power generator to a propulsorto drive the propulsor, the power system being operable in a plurality of machine modes, each machine mode corresponding to respective configurations of the power system components such that a type of power is delivered from at least one power generator to at least one propulsor,the system being configured to:obtain power demand data indicating a current power demand of a marine vessel to achieve a vessel objective; anddetermine, for each one of a plurality of machine modes and for a plurality of power output configurations for each machine mode, a specific fuel oil consumption for the vessel to achieve the objective and for the vessel power system to meet the power demand, wherein each power output configuration for each machine mode corresponds to respective power output levels of each power generator configured to deliver power to a propulsor inthat machine mode, the sum of the power output levels for each power generator in each power output configuration being equal to the current power demand.

26. The system of claim 25, the system being further configured to:transmit the machine mode and power output configuration that is associated with the lowest specific fuel oil consumption to a vessel power system control unit.

27. The system of claim 25 or 26, the system being further configured to:cause the power system to be operated in the machine mode and according to the power output configuration associated with the lower specific fuel oil consumption.

28. The system of any of claims 25-27, the system being further configured to:obtain vessel objective data indicating the objective that the vessel is to achieve.

29. The system of any of claims 25-28, the system being further configured to:obtain operational data indicating the machine modes in which the power system of the vessel can be configured to operate.

30. The system of any of claims 25-29, wherein the objective that the vessel is to achieve is associated with the vessel operating in a transit mode.

31. The system of any of claims 25-30, the system being further configured to:obtain input indicating that the vessel is to be operating, or is operating, in a transit mode.

32. The system of any of claim 25-31 wherein determining the specific fuel oil consumption, for each one of a plurality of machine modes and for a plurality of power output configurations for each machine mode, is based on a speed of the vessel.

33. The system of claim 32, the system being further configured to:obtain data indicating the speed of the vessel.

34. The system of any of claims 25-33, the system being further configured to:determine, for each machine mode and for each power output configuration, the specific fuel oil consumption for the vessel to achieve the objective and to meet the power demand for a plurality of different vessel speeds.

35. The system of claim 34, the system being further configured to:transmit the machine mode and power output configuration and vessel speed that is associated with the lowest specific fuel oil consumption to a vessel power system control unit.

36. The system of claim 33 or 34, the system being further configured to:cause the power system to be operated in the machine mode and according to the power output configuration, to thereby cause the vessel to be operated at the associated vessel speed, that is associated with the lowest specific fuel oil consumption.

37. The system of any of claims 25-36, the system being further configured to:determine, for each machine mode and for each power output configuration, the specific fuel oil consumption by:causing the current power demand to be input into a machine learning model trained based on historic data relating to specific fuel oil consumptions for a plurality of power output configurations and for a plurality of machine modes to output at least one specific fuel oil consumption associated with operating the vessel power system in at least one machine mode according to at least one power output configuration to meet the current power demand of the vessel based on receiving the current power demand of the vessel as input; andobtaining the specific fuel oil consumption output by the machine learning model.

38. The system of any of claims 25-37, the system being further configured to:determine, for each machine mode and for each power output configuration, the specific fuel oil consumption by:causing the current power demand of the vessel to be input into a plurality of machine learning models, each machine learning model in the plurality being trained based on historic data relating to specific fuel oil consumptions for a plurality of power output configurations in a respective machine mode to output a plurality of specific fuel oil consumptions associated with operating the vessel power system in the respective machine mode according to the plurality of power output configurations to meet the current power demand of the vessel based on receiving the current power demand of the vessel as input; andobtaining the specific fuel oil consumption output of each machine learning model for each machine mode.

39. The system of claim 37 or 38, wherein the or each machine learning model is trained based on historic data relating to specific fuel oil consumptions for a plurality of power output configurations and for the or each machine mode for the vessel operating in a transit mode.

40. The system of any of claims 25-39 wherein the plurality of machine modes comprises at least one of:a diesel mechanic mode in which at least one power generator is configured to generate mechanical power to power at least one propulsor,a diesel electric mode in which at least one power generator is configured to generate electrical power to power at least one propulsor, anda hybrid mode in which at least one power generator is configured to generate mechanical power to power at least one propulsor at least one power generator is configured to generate electrical power to power at least one propulsor.

41. The system of any of claims 25-40, wherein the power generators of the vessel power system comprise one or more of: a main engine configured to generate mechanical power, an auxiliary engine configured to generate electrical power, a generator configured to generate electrical power, and a motor configured to generate mechanical power.

42. The system of any of claims 25-41 , the system being further configured to:obtain vessel operational data, wherein the specific fuel oil consumption is determined for each machine mode and for each power output configuration based on the vessel operational data.

43. The system of claim 42, wherein the vessel operational data comprises at least one of: information relating to at least one component of the vessel power system and the flow rate of fuel to at least one power generator.

44. The system of any of claims 25-43, the system being further configured to:obtain data indicating at least one of: the vessel speed, an operational state of at least one component of the vessel power system, and the flow rate of fuel flow to at least one power system;determine a machine mode in which the vessel is operating;determine a number of power generators of the vessel power system that are running according to respective power output levels;determine the specific fuel oil consumption of the vessel; andstore the specific fuel oil consumption, the power output configuration being the respective power output levels of the power generators, a route or a plurality of routes the vessel could traverse, a weather condition or set of weather conditions the vessel could be subject to while traversing a given route, power demand data as described above, an emission (e.g. greenhouse-gas or carbon) emission for the vessel while traversing a given route, and the machine mode, as training data.

45. The system of any of claims 25-44 wherein the power demand data comprises a power demand for a vessel to traverse a given route in a given weather condition, the weather condition being based on weather data.

46. The system of any of claims 25-45 wherein the power demand data comprises a total non-propulsion load data comprising an determined power required to operate at least one non-propulsion load of the vessel along a given route.

47. The system of any of claims 25-46 wherein the determined SFOC is based on an emissions value for a given route the vessel is to traverse, the determined SFOC being based on a route associated with an emissions value that is less than a predetermined threshold.

48. A cloud computing environment system comprising the system of any of claims 25-47.

49. Computer hardware configured to implement the method of any of claims 1 -24, the system of any of claims 25-47, or the cloud computing environment system of claim 48.

Images