Optimization strategy for hybrid energy storage system with intelligent adaptive control

Abstract

The present invention discloses a control system for implementing a hybrid energy storage system (ESS) optimization strategy. The hybrid ESS optimization strategy can be implemented in a machine including a power system and a power controller, the power system including multiple power sources, and the power controller including one or more processors. The power controller can receive information related to a set of brake fuel consumption (BSFC) maps associated with the multiple power sources, determine performance indicators using cost functions associated with the multiple power sources, and generate commands to operate the power system based on a power distribution that minimizes energy costs, so as to operate the power system based on information related to the set of BSFC maps, the performance indicators, and the load associated with the power system.

Description

Technical Field

[0001] The present disclosure relates generally to powertrain management, and more particularly to a control system that can implement a hybrid energy storage system (ESS) optimization strategy with intelligent adaptive control. Background Art

[0002] Mobile machines such as ships typically include a power system in which multiple power sources (e.g., one or more engines, one or more batteries, etc.) are wired together to drive one or more main loads (e.g., propulsion) and one or more auxiliary loads (e.g., heating, ventilation and air conditioning (HVAC), lighting, pumps, etc.). In many cases, the multiple power sources may have different operating ranges, speeds, efficiencies, expected service lives, etc. The power sources may be mechanically connected to the loads or electrically connected to the loads through generators. In some applications, the loads of the machine may be driven both mechanically and electrically in a hybrid arrangement. For example, a hybrid system may include one or more combustion engines (e.g., gasoline engines, diesel engines, dynamic gas hybrid (DGB) engines) and an energy storage system (ESS), which may use energy captured at an earlier time to process at least a portion of the load associated with the machine in order to optimize fuel consumption, reduce emissions, improve the transient response performance of the system, perform peak load suppression to maintain system stability, reduce overall ownership and operating costs, and the like. However, due to the variability and fluctuation of the load (s) to be driven by the power system, the different capabilities of the multiple power sources, and / or other dynamic variables, determining the optimal distribution of power demand between the multiple power sources may be a challenging task.

[0003] An attempt to allocate power and energy between different power sources is disclosed in U.S. Patent No. 10,170,912 (the "'912 Patent"), issued on January 1, 2019 to Mathiesen et al. In particular, the '912 Patent discloses a vessel having an energy control system (ECS) that automatically adapts to the power plant configuration and the type of motor generator set (MGS) and energy storage unit (e.g., battery or capacitor), and calculates a ramp (kW / s) of the total available load at any particular moment. The calculated total load ramp value is sent to a dynamic positioning (DP) control system and is used to coordinate changes in thruster command signals provided by thrust distribution optimization (212) with changes in the maximum load ramp signal. The '912 Patent also discloses that the system automatically accounts for changes in dynamic or available instantaneous power capacity based on the characteristics of the connected MGS and battery units, as the dynamic capacity of the battery may vary significantly based on the charge / discharge rate of the battery.

[0004] While the ECS described in the '912 patent can automatically adapt to the power plant configuration and the type of MSG and energy storage unit (e.g., battery or capacitor) to adjust and calculate the gradual changes in the total amount of available load of the ship's power plant at any particular moment, the '912 patent fails to disclose technology for determining a dynamic optimization map that provides the best power distribution for the power sources included in the ship's power plant based on different operating conditions, information related to the health and / or remaining useful life of the power sources, and / or other dynamic variables.

[0005] The hybrid ESS optimization control system of the present disclosure solves one or more of the above-mentioned problems and / or other problems in the prior art. Summary of the invention

[0006] According to one embodiment, a method may include: receiving information relating to a load associated with a power system and information relating to a set of brake specific fuel consumption (BSFC) maps associated with the power system; determining a performance indicator using a cost function associated with multiple power sources included in the power system; determining a power distribution among the multiple power sources based on the set of BSFC maps and the performance indicator to minimize the energy cost associated with the load; and generating a command to operate the power system based on the power distribution, wherein the command indicates a first amount of power associated with operating one or more engines among the multiple power sources, and a second amount of power associated with operating one or more batteries among the multiple power sources.

[0007] According to another embodiment, a device may include one or more memories and one or more processors, wherein the one or more processors are communicatively coupled to the one or more memories and are configured to: receive a set of input parameters including a load associated with a power system, a plurality of BSFC maps respectively associated with a plurality of engines, and a battery equivalent BSFC map associated with one or more batteries; determine one or more performance indicators using a cost function based on health information associated with the plurality of engines and health information associated with the one or more batteries; generate a dynamic optimization map based on the set of input parameters and the one or more performance indicators; and generate a command based on the dynamic optimization map to operate the power system according to a power distribution that minimizes the energy cost associated with the load.

[0008] According to yet another embodiment, a machine may include: a power system including a plurality of power sources; and a power controller including one or more processors. The one or more processors may be configured to: receive information related to a set of BSFC maps associated with the plurality of power sources; determine a performance indicator using a cost function based on health information associated with the plurality of power sources; and generate a command to operate the power system based on a power distribution that minimizes energy cost, so as to operate the power system based on information related to the set of BSFC maps, the performance indicator, and a load associated with the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1 is a diagram of one or more exemplary embodiments of a control system that may implement the hybrid energy storage system (ESS) optimization strategy with intelligent adaptive control described herein.

[0010] Figure 2 is a diagram of one or more exemplary implementations of a control system that may implement the hybrid ESS optimization strategy with intelligent adaptive control described herein.

[0011] Figure 3 is a flow chart of an exemplary process for implementing a hybrid ESS optimization strategy with intelligent adaptive control.

[0012] Figure 4 is a diagram of an exemplary dynamic optimization map that may be used in one or more embodiments of the hybrid ESS optimization strategy with intelligent adaptive control described herein.

[0013] Figure 5 is a graph of one or more exemplary implementations of a hybrid ESS optimization strategy with intelligent adaptive control as described herein.

[0014] Figure 6 is a graph of one or more exemplary implementations of a hybrid ESS optimization strategy with intelligent adaptive control as described herein.

[0015] Figure 7 is a graph of one or more exemplary implementations of a hybrid ESS optimization strategy with intelligent adaptive control as described herein.

[0016] Figure 8 is a flow chart of an exemplary process for implementing a hybrid ESS optimization strategy with intelligent adaptive control. DETAILED DESCRIPTION

[0017] The present disclosure relates to a power controller that can implement a hybrid energy storage system (ESS) optimization strategy with intelligent adaptive control. The hybrid ESS optimization strategy has general applicability to any machine having a power system that includes multiple power sources, such as a generator set (or "genset"), including a combustion engine, a generator, etc., ESS elements, such as batteries, etc. As some examples, the machine can be a ship, an aircraft, a vehicle, etc. In addition, one or more appliances can be connected to the machine and driven from the power system based on one or more control signals generated using a power controller that implements a hybrid ESS optimization strategy with intelligent adaptive control.

[0018] Figure 1 1 is a diagram of one or more exemplary embodiments 100 of a control system that may implement a hybrid ESS optimization strategy with intelligent adaptive control. Figure 1 As shown in , (multiple) embodiments 100 may include a power controller 110 configured to control a power system including multiple power sources. Figure 1 As shown in , the plurality of power sources may include a generator set 120 (sometimes referred to as a "genset"), which includes one or more engines 122 (shown as engines 122-1 to 122-N, where N is an integer and N ≥ 1) connected to corresponding generators and engine controllers. In addition, the plurality of power sources may include an ESS 130, which may include one or more devices configured to store energy, which may be released or otherwise used at a later time. For example, the ESS 130 may include one or more batteries connected to a controller device and a bidirectional power (BDP) inverter that converts direct current into alternating current. As described herein, the plurality of power sources may be configured to selectively supply energy to support a load 140. For example, the engine(s) 122 may be operably connected to the ESS 130 and the load 140 to selectively supply power thereto, and the ESS 130 may be operably connected to the engine(s) 122 to selectively receive and supply power relative to the engine 122, and may be operably connected to the load 140 to selectively supply power to support the load 140. The power controller 110 and the power system may be operably associated with a machine, such as a vessel, a locomotive, an on- or off-highway vehicle, a generator set, and / or another machine that performs work.

[0019] The (multiple) engines 122 may include any number and / or type of combustion engines, some or all of which may be connected to corresponding generators to form a generator set. For example, the (multiple) engines 122 may include gasoline or diesel combustion engines, dynamic gas hybrid (DGB) engines, fuel cells, natural gas engines, etc. The (multiple) engines 122 may also include a prime mover (e.g., a crankshaft) that is mechanically decoupled from the ESS 130 and the load 140. The (multiple) engines 122 may be operably connected to the ESS 130 and the load 140 via one or more electrical circuits and / or hydraulic circuits, and may effectively operate as an energy source or power source. The engine 122 may be controlled by an engine control module (ECM) to selectively generate a varying amount of power based on the ECM changing one or more engine operating parameters of the engine 122. For example, the engine operating parameters may include the amount of fuel injected into the combustion chamber, the opening and / or closing timing of the intake and / or exhaust valves, etc. The mechanical output of engine(s) 122 may be routed directly to load 140 (e.g., mechanically to a drive shaft, such as a propeller drive shaft for a marine vessel) and / or indirectly to a drive shaft via a generator (e.g., electrically to motors for propellers and other auxiliary loads).

[0020] In some embodiments, multiple engines 122 can cooperate to meet the power demand determined by the load 140. For example, a larger medium-speed engine 122 can be able to achieve greater power output at higher fuel efficiency (e.g., lower fuel consumption) and / or lower emissions. In contrast, a smaller high-speed engine 122 can be able to achieve faster transient response and high-efficiency operation for machines operating at lower loads. A particular machine may include multiple engines 122 with different capabilities, etc. Additionally or alternatively, a particular machine may include multiple engines 122 with the same rating and efficiency but with different capabilities, etc.

[0021] The ESS 130 may include one or more devices that can store previously generated energy for subsequent use. For example, the ESS 130 may include one or more batteries, batteries, capacitors, and / or other suitable devices that can store and subsequently release energy. In some embodiments, the devices that make up the ESS 130 may be arranged to be charged by energy received from (multiple) engines 122, and may subsequently release energy. The released energy may provide a backup energy source for the engine 122, may provide peak load suppression by acting as a buffer and / or a leveling agent for the load 140 driven by the engine 122, may provide instantaneous power to support the load 140, and the like. The chemical energy generated by burning the fuel may be converted into kinetic energy and thermal energy. The kinetic energy may cause the machine to rotate. The generator may convert the rotation into electrical energy (e.g., current and / or voltage), and the electrical energy may be stored in the ESS 130 by charging the ESS 130.

[0022] Loads 140 may include profiles associated with electrical requirements for propulsion machinery, such as electrical requirements for powering an electrically driven propulsion unit that can independently rotate relative to the hull of the vessel to propel the vessel. Additionally or alternatively, loads 140 may include other electrical requirements of the machinery, such as electrical requirements for operating auxiliary loads, such as appliances for lifting and moving freight, communications equipment, heating, ventilation, and air conditioning (HVAC) systems, lighting, etc.

[0023] like Figure 1 As further shown in , a set of input parameters 150 can be provided to a cost determination unit 160. The cost determination unit 160 may include one or more processors configured to execute a cost function to generate one or more performance indicators. For example, the power controller 110 may evaluate (multiple) performance indicators based on the current load 140 to determine the power distribution (or load sharing) ratio of each power source included in the power system. For example, as described herein, the power controller 110 may identify a number of power sources in a given power system and the capacity of each power source, identify a given power load, identify an operation to be performed on the given power load, identify the total amount of energy (e.g., in kilowatts (kW)) to be generated by the power system in order to perform the desired operation on the given load, generate a control signal or command indicating a desired ratio of the total amount of energy generated by each power source included in the power system, and so on. Therefore, each power source included in the power system can generate a mechanical and / or electrical power output based on the desired ratio of the total amount of energy to be generated by each corresponding power source.

[0024] like Figure 1As further shown in FIG. 1 , the set of input parameters 150 provided to the cost determination unit 160 may include asset status information 152, asset priorities 154, asset constraints 156, brake specific fuel consumption (BSFC) maps 158, etc. The asset status information 152 may indicate whether a given engine 122 and / or device in the ESS 130 is online, offline, idle, and ready for a quick ramp up, etc. In this way, the asset status information 152 may indicate which power source in the power system is ready and available for use, how quickly an offline or idle power source can be brought online and available, etc., which may enable the power controller 110 to determine whether to turn on or off an engine 122, a battery, etc. based on the incoming load 140.

[0025] The asset priority 154 may include priority information associated with each engine 122 and / or device in the ESS 130, which may vary based on the load 140, operating mode, health information, etc. For example, the asset priority 154 may vary based on the size of the engine 122, the fuel source of the engine 122, the capacity or state of charge (SOC) level of the battery in the ESS 130, etc. The priority may be affected by maintenance time, asset health, asset efficiency, etc. In addition, the priority may vary based on whether the machine is operating in a high performance mode, a fuel economy mode, an emission reduction mode, etc. For example, if the load 140 is associated with a fast response demand, the asset priority 154 may assign the highest priority to a battery that can provide instantaneous power to support the engine 122, and an engine 122 that can provide a fast transient response time (e.g., a turbocharged diesel engine) may be assigned a higher priority than another engine 122 with a slower transient response time (e.g., a gasoline engine).

[0026] Additionally, in some embodiments, the asset priorities 154 may be customer defined (e.g., based on the performance of different engines 122 produced by different manufacturers) and / or health information to maximize or extend the useful life of the power source. For example, if two engines 122 have the same or substantially similar performance ratings, then the engine 122 that is less used or closer to the end of its useful life may be assigned a higher priority (e.g., to balance the number of operating hours between the engines 122).

[0027] The asset constraints 156 may include information related to one or more limitations on the corresponding power source. For example, the asset constraints 156 may include information related to the minimum and / or maximum amount of power to be generated by a particular power source. In another example, the asset constraints 156 may integrate health information associated with the engine 122 and / or battery included in the ESS 130. For example, the asset constraints 156 may indicate that an engine 122 with a diagnostic problem or requiring service is to be placed offline or otherwise made unavailable. Additionally or alternatively, the asset constraints 156 may include information related to the minimum and / or maximum duration that the engine 122 is in an on or off state, information to prevent the battery from being overcharged or used in a manner that may cause component degradation, and the like.

[0028] The BSFC map 158 may include a two-dimensional curve and / or a three-dimensional map that measures the efficiency of each power source. For example, the BSFC map 158 may include one or more BSFC maps that measure the fuel efficiency of the engine 122. Additionally or alternatively, one or more battery equivalent BSFC maps for batteries or other devices included in the ESS 130 may be based on the cumulative energy cost associated with the charging and discharging of one or more batteries, the estimated life associated with one or more batteries, and / or the replacement cost of one or more batteries. In some embodiments, each BSFC map 158 may be a two-dimensional curve that measures the fuel consumption rate of the corresponding power source as a function of the generated power. Additionally or alternatively, the BSFC map 158 may be based on one or more functions of both the generated power and the speed of operating the corresponding power source, and thus may include a three-dimensional map. The BSFC map 158 may be different for each engine 122 and / or for each different type of power source. In some embodiments, the BSFC map 158 may be input to the cost determination unit 160 to determine the combined fuel consumption rate and / or the generated power of all power sources under different possible load sharing configurations.

[0029] like Figure 1 As shown in , asset status information 152, asset priorities 154, asset constraints 156, and BSFC maps 158 can be input as parameters to a cost determination unit 160, which can map the values ​​of the various parameters to outputs (e.g., real numbers) representing costs associated with the mapped values. For example, the output from the cost determination unit 160 can represent the cost of a specific load sharing or power distribution among the power sources included in the power system. Therefore, the output from the cost determination unit 160 can be provided to the power controller 110, which can include an optimization component configured to select a specific power distribution that minimizes a target variable, such as fuel consumption, emissions, transient performance loss, etc.

[0030] For example, the optimization component may select a particular power distribution configuration that minimizes the financial cost per unit of energy produced by the power system (e.g., dollars per kilowatt-hour ($ / kWh)). Thus, the power controller 110 may issue a command that causes each power source in the power system to generate a corresponding amount of energy (e.g., in kilowatts) specified by the optimization component. For example, the command may indicate an amount of power to be generated by the engine 122, which may be evenly or unevenly distributed among the engines 122 based on health information associated with the engines 122. Additionally or alternatively, the command may indicate an amount of power associated with operating one or more batteries or other devices included in the ESS 130, wherein the amount of power associated with operating one or more batteries may, for example, have a negative value to indicate that the battery is to be charged by operating the engine 122, or have a positive value to indicate that the battery is to be discharged to supply power, support or replace the engine 122.

[0031] As mentioned above, providing Figure 1 As one or more examples. Other examples can be combined with Figure 1 Different than described.

[0032] Figure 2 is a diagram of one or more exemplary embodiments 200 of a control system that may implement the hybrid ESS optimization strategy with intelligent adaptive control described herein. (Multiple) exemplary embodiments 200 may use one or more engine BSFC maps 158-1 in combination with one or more battery equivalent BSFC maps 158-2 as inputs to a cost determination unit 160 to determine an appropriate power allocation ratio for using one or more batteries as a power source for a machine. For example, as described above, the engine BSFC map 158-1 may generally measure fuel efficiency (e.g., in grams per kWh (g / kWh)) for one or more corresponding engines as a function of the power generated by the one or more engines. Therefore, because batteries do not consume fuel, the battery equivalent BSFC map 158-2 may provide a mechanism to integrate the efficiency of one or more batteries into the cost determination unit 160.

[0033] For example, the battery equivalent BSFC map 158-2 may generally provide a constant BSFC regardless of the power produced by one or more batteries because fuel consumption does not vary at different power levels. Therefore, because the engine may generally consume fuel while charging the batteries, the constant BSFC provided in the battery equivalent BSFC map 158-2 may be based on the fuel consumed by the engine while charging the batteries. For example, if the engine is run for a period of time to charge the batteries, the constant BSFC may be based on the fuel consumed by the engine during the period of time when the engine is running to charge the batteries, and no fuel cost may be associated with any period of time when the batteries are discharged and / or charged without using power from the engine (e.g., using braking energy in a hybrid vehicle, using water-induced propeller motion in a ship, etc.).

[0034] Thus, in some embodiments, the battery capacity integrator 210 may receive a command or control signal indicating the amount of power associated with operating one or more batteries, which may, for example, have a positive value to indicate that the battery is being charged by running the engine, or a negative value to indicate that the battery is to be discharged. Typically, when the battery is discharged, no fuel cost is incurred, but each time the engine is run to charge the battery, the engine may consume a certain amount of fuel. Therefore, the battery capacity integrator 210 may measure the fuel consumed during the (multiple) time periods in which the engine is operated to charge the battery, and integrate the fuel consumption into the battery equivalent BSFC map 158-2. In some embodiments, the battery capacity integrator 210 may determine an average value of the fuel consumed to charge the battery based on the load 140, the amount of time the engine is run to charge the battery, etc., which average value may be used as a constant BSFC. In addition, each time the engine is run to charge the battery, the battery capacity integrator 210 may update the constant BSFC to represent the average fuel consumed to charge the battery to a level sufficient to produce the required amount of energy, the bidirectional efficiency loss associated with battery charging and discharging, etc.

[0035] like Figure 2As further shown in , the cycle and operating mode control logic 220 may receive information related to the current load request 140 and / or a set of input parameters 150, which may include asset status information 152, asset priorities 154, asset constraints 156, BSFC maps 158, etc. In addition, the set of input parameters 150 provided to the cycle and operating mode control logic 220 may include information related to the current operating mode, such as fuel economy, emission reduction, performance, spinning reserve, peak load suppression, energy harvesting, backup power, etc. Therefore, as shown by reference numeral 225, the cycle and operating mode control logic 220 may apply one or more machine learning techniques to learn and adapt to input parameters, such as the current load request 140, historical load information, predicted upcoming load, battery state of charge (SOC) level, battery charging and / or discharging cycles, etc. Therefore, as described herein, the cycle and operating mode control logic 220 may apply machine learning techniques to implement an informed decision process when determining one or more performance indicators that may serve as inputs to the cost determination unit 160 to further optimize the power profile output by the power controller 110.

[0036] For example, in some embodiments, the loop and operating mode control logic 220 can train a machine learning model (e.g., a multi-engine optimization model) using a supervised training procedure. The inputs to the machine learning model can be specified by a subject matter professional, which can reduce the amount of time required to train the model, the amount of processing resources, etc. relative to an unsupervised training procedure. In some embodiments, the loop and operating mode control logic 220 can use one or more other model training techniques, such as neural network techniques, latent semantic indexing techniques, etc. For example, the loop and operating mode control logic 220 can perform artificial neural network processing techniques (e.g., using a two-layer feedforward neural network architecture, a three-layer feedforward neural network architecture, etc.) to perform pattern recognition regarding costs associated with certain patterns of input parameters.

[0037] For example, supervised multi-label classification techniques may be used to train the model.For example, as a first step, the loop and operating mode control logic 220 may map historical information related to the load request 140, input parameters 150, operating mode, etc. to different costs.

[0038] As a second step, the loop and operating mode control logic 220 can determine a chain of classifiers, thereby associating labels of target variables. In this case, the loop and operating mode control logic 220 can use the output of the first label as the input of the second label, and can determine the likelihood that a particular load request 140, input parameter 150, operating mode, etc. is associated with a particular cost based on similarity to other parameters including similar characteristics. In this way, the loop and operating mode control logic 220 converts the classification from a multi-label classification problem to multiple single-classification problems, thereby reducing processing utilization.

[0039] As a third step, the loop and operating mode control logic 220 may determine a Hamming loss metric related to the accuracy of the label in performing the classification by using a validation set of the data. For example, applying a weighting to each parameter and the accuracy of whether each parameter is associated with a particular cost results in correctly predicting whether a given power profile is optimal, thereby taking into account the different amounts by which the association of any one parameter affects the power profile.

[0040] As a fourth step, the loop and operating mode control logic 220 can finalize the model based on labels that meet a threshold accuracy associated with the Hamming loss metric, and can use the model to next predict the power profile to be used when certain combinations of load requests 140, input parameters 150, operating modes, etc. are received.

[0041] In some embodiments, a different device, such as a server device, can generate and train the model. The different device can send the model for use by the cycle and operating mode control logic 220. The different device can update and send the model to the cycle and operating mode control logic 220 (e.g., based on a schedule, on demand, based on a trigger, based on a periodicity, etc.).

[0042] Thus, the cycling and operating mode control logic 220 may use any number of artificial intelligence techniques, machine learning techniques, deep learning techniques, etc. to learn and adapt to different load requests 140, input parameters 150, operating modes, etc. Additionally, in some embodiments, the battery capacity integrator 210 may use similar artificial intelligence techniques, machine learning techniques, deep learning techniques, etc. to determine the fuel cost associated with battery charging.

[0043] As mentioned above, providing Figure 2 As one or more examples. Other examples can be combined with Figure 2 Different than described.

[0044] Figure 3 is a flow chart of an exemplary process 300 for implementing a hybrid ESS optimization strategy with intelligent adaptive control. Figure 3One or more process blocks may be performed by a device, such as a power controller (e.g., power controller 110) and / or another device or group of devices separate from or including the power controller (e.g., an ECM associated with engine(s) 122, a BDP inverter or controller associated with ESS 130, and / or another device associated with embodiment(s) 100, embodiment 200, etc.).

[0045] like Figure 3 As shown in , process 300 may include generating a dynamic optimization map indicating optimal conditions for charging and / or discharging one or more batteries (block 310). For example, as described in more detail elsewhere herein, the dynamic optimization map may include a two-dimensional map or a three-dimensional map to indicate potential fuel savings as a function of various parameters, such as the charge level associated with one or more batteries (e.g., total power before charging and / or total power after charging), the load to be driven by the power system, etc. In some embodiments, when the power system is in an offline state, the power controller may initially generate a dynamic optimization based on a set of default or baseline parameters, such as a baseline BSFC curve associated with one or more engines, a baseline battery equivalent BSFC curve for charging one or more batteries based on typical fuel consumption parameters, etc.

[0046] like Figure 3 As further shown in FIG. 3 , process 300 may include receiving a load profile input (block 320). For example, as described above, the load profile input may include information related to primary loads including electrical requirements of a propulsion machine, such as electrical requirements for powering an electrically driven propulsion unit that may independently rotate relative to a hull of the vessel to propel the vessel. Additionally or alternatively, the load profile may include information related to one or more auxiliary loads, such as electrical requirements for operating appliances for lifting and moving freight, communications equipment, heating, ventilation, and air conditioning (HVAC) systems, lighting, and the like.

[0047] like Figure 3 As further shown in FIG. 3 , process 300 may include obtaining a BSFC map associated with one or more engines (block 330). For example, a BSFC map obtained from an engine(s) Figure 1 The fuel efficiency of a corresponding engine(s) may generally be measured as a function of the power produced by one or more engines.

[0048] Next, process 300 may include obtaining one or more battery equivalent BSFC maps (box 340). For example, the battery equivalent BSFC map may provide a constant BSFC regardless of the power produced by one or more batteries because the fuel consumption does not change at different power levels. Therefore, the battery capacity integrator may measure the fuel consumed by one or more engines during the time period when the (multiple) engines are operated to charge the batteries. In some embodiments, the battery capacity integrator may use one or more artificial intelligence techniques, machine learning techniques, deep learning techniques, etc. to learn and adapt parameters related to the fuel consumed to charge the batteries. In addition, in some embodiments, process 300 may include inputting information related to the expected service life of one or more batteries, the replacement cost of one or more batteries, etc., as a penalty factor that can be considered when calculating the battery equivalent BSFC map (box 350).

[0049] like Figure 3 As further shown in FIG. 3 , process 300 may include identifying any location-specific restrictions based on the current position or orientation of a machine associated with the power system (block 360). For example, in some cases, the machine may be a vessel that may be operated in a dynamic positioning zone, which generally refers to an area in which a vessel employs a computer-controlled dynamic positioning (DP) system to automatically maintain the position and heading of the vessel using onboard propellers and thrusters. In such cases, the DP system may use a mathematical model including information related to the current position and orientation of the vessel and information related to environmental forces that affect the position and / or orientation of the vessel to calculate the required steering angle and thruster output of each thruster, which may allow operation in seas where mooring or anchoring may not otherwise be feasible. Therefore, since a vessel operating in a dynamic positioning zone may need to have a highly reliable power system to prevent deviation from the course, enable maneuvering in a degraded state, etc., a dynamic positioning zone may be subject to the requirement that a backup power source be available in the event of an interruption or other failure of the primary power source. Therefore, in some embodiments, a power controller may receive information relating to the current position of a machine associated with the power system, and determine whether there are any location-specific restrictions or other requirements for operating the power system in the current position of the machine.

[0050] Process 300 may include updating a dynamic optimization map that indicates optimal conditions for charging and discharging one or more batteries (block 370). For example, the dynamic optimization map may be updated based on operating conditions, such as load profile inputs, BSFC map(s) obtained from engine(s), battery equivalent BSFC map(s), any location-specific constraints, etc. In some embodiments, the updated dynamic optimization map may be used by the controller to determine a power distribution ratio to be allocated between the (individual and / or collective) engines and batteries. Additionally or alternatively, the power controller may calculate the instantaneous power distribution based on operating conditions using a real-time optimization function.

[0051] Process 300 may further include generating a power profile and battery charge / discharge commands based on current operating conditions (block 380). In some embodiments, the power controller may generate the power profile and battery charge / discharge commands using a current dynamic optimization map. Additionally or alternatively, the power controller may recalibrate the instantaneous power profile based on the current operating mode of the power system using a real-time optimization function based on operating conditions.

[0052] For example, in the case where the power system is operated in a transient mode requiring a fast response time, the power controller may use a dynamic optimization map to generate a power distribution and battery charge / discharge commands, because the real-time optimization function may not converge quickly enough to meet the response time required for the transient mode. In other cases, such as when the power system is operated in a fuel economy mode, an emission reduction mode, a high-performance mode, etc., the power controller may use a real-time optimization function to generate a power distribution and a battery charge / discharge command to ensure that each power source included in the power system operates at maximum efficiency.

[0053] In some embodiments, the power distribution and battery charge / discharge commands may generally indicate the amount of power (e.g., kilowatts) generated by each engine included in the power system, and the amount of power (e.g., kilowatts) received and / or generated by each battery included in the power system. For example, as described above, the power controller may determine the optimal power distribution between the (multiple) engine and (multiple) battery power sources (e.g., based on a dynamic optimization map or instantaneous calculation) taking into account the current load, operating mode, health information associated with the power source, replacement cost, service cost, fuel cost, the region where the machine is located, etc.

[0054] For example, in the case where the machine is in a dynamic positioning zone requiring a backup power source, the power profile and battery charge / discharge commands can ensure that at least two power sources are operable (e.g., multiple engines, at least one engine and at least one battery, etc.). In another example, when there is a relatively small load (e.g., when the load fails to meet a load threshold) and / or the battery has a low charge level (e.g., when the charge level fails to meet a charge threshold), the commands can cause one or more engines to generate a certain amount of power to charge one or more batteries.

[0055] In other examples, the commands may cause the battery to discharge to perform peak load suppression to act as a buffer and prevent the engine from having to support a load above a load threshold, provide immediate or transient power to support a generator driven by the engine, etc. Additionally or alternatively, the commands may cause the battery to charge to harvest or otherwise recover energy generated by a renewable energy source or a machine such as a lift or drilling equipment, provide spinning reserve as a backup for an engine driven generator and / or reduce the number of online engines, etc.

[0056] like Figure 3 As further shown in FIG. 3 , process 300 may include calculating the energy cost of cumulative battery charging and discharging based on the power profile and the charge / discharge commands (block 390). For example, as described above, in some cases, the battery may be charged by running the engine for a period of time, which may cause the engine to consume a certain amount of fuel. Thus, as Figure 3 As shown in , the accumulated fuel consumption cost for charging the battery can be integrated into a battery equivalent BSFC map, which is used to update the dynamic optimization map using a real-time optimization function and / or calculate the instantaneous power distribution based on operating conditions.

[0057] Process 300 may include combining Figure 3 Variations of those embodiments described and / or additional embodiments, such as any single implementation or any combination of embodiments described elsewhere herein. Figure 3 An exemplary block diagram of process 300 is shown, but in some examples, Figure 3 The process 300 may include additional blocks, fewer blocks, different blocks, or differently arranged blocks than those depicted in FIG. Additionally or alternatively, two or more of the blocks in the process 300 may be performed in parallel.

[0058] Figure 4is a diagram of an exemplary dynamic optimization map 400 that may be used in one or more embodiments of the hybrid ESS optimization strategy with intelligent adaptive control described herein. For example, as described above, the dynamic optimization map 400 may be used to determine the power distribution ratio of the power system (e.g., the percentage of the load to be distributed between various power sources, such as one or more engines and / or batteries) in situations where a fast response time is required or where instantaneous calculations cannot be performed for some reason when generating commands for operating the power system.

[0059] like Figure 4 As shown in FIG. 4 , the dynamic optimization map 400 can indicate potential fuel savings (e.g., expressed as a percentage) as a function of total power before charging one or more batteries and total power before discharging one or more batteries. In addition, in some embodiments, the dynamic optimization map 400 can integrate information related to different load parameters. Thus, given a particular load to be handled by the powertrain system, the dynamic optimization map 400 can indicate potential fuel savings that can be achieved by using the engine to charge the battery while the engine is handling the load, rather than discharging the battery to handle the load or to provide support for the engine.

[0060] For example, as shown at reference numeral 410, the battery may be discharged to support a high load and delay or avoid starting additional engines (e.g., reducing the number of engines online to support the load when the engines online cannot handle the full load). In this way, the battery may be discharged to handle at least a portion of the load, which may provide fuel savings due to a reduction in the number of engines being operated.

[0061] In another example, as shown at reference numeral 420, a low load condition may provide an opportunity to drive one or more engines to operate at a high speed, which provides an optimal efficient point for charging one or more batteries. For example, an engine may generally have more efficient fuel consumption at higher speeds, so a low load condition may provide an opportunity to operate the engine at a high (fuel efficient) speed to generate excess power (e.g., more power than required to handle the load), thereby charging the batteries at a lower fuel cost.

[0062] As mentioned above, providing Figure 4 As one or more examples. Other examples can be combined with Figure 4 Different than described.

[0063] Figure 5500 is a graph of one or more exemplary implementations of a hybrid ESS optimization strategy with intelligent adaptive control as described herein. For example, as described above, graph 500 may represent a real-time instantaneous calculation of the total load in the power system based on the individual loads supported by the multiple engines (shown as engines 1-4) and the battery included in the power system. Figure 5 In FIG. 1 , the x-axis represents time and the y-axis represents load (eg, in kilowatts).

[0064] like Figure 5 As shown in , the power controller may initially operate only the first engine (engine 1) during the time period when the first engine is able to handle the total load. For example, the total load experiences a small increase in the first shaded vertical bar, and the first engine is pushed to a higher load to support the increase in the total load. Therefore, as shown in reference numeral 510, the power controller may delay starting the second engine based on the graph 500 indicating that the first engine is able to handle the load increase until the total load increases to a point beyond which the first engine is able to handle the load independently.

[0065] In this way, by delaying the start of the second engine, fuel can be saved during the time period in which the second engine is allowed to remain offline. In addition, as shown in reference numeral 520, the power controller can completely avoid starting the third engine because the first engine and the second engine can handle the total load. In this way, additional fuel can be saved by reducing the number of engines running to support the load and / or reducing the amount of time the engine runs. For example, running an engine at a low load level may generally be less fuel efficient than running an engine at a high load level. Therefore, by running the first engine at a high load and delaying the start of the second engine until the total load exceeds the capacity of the first engine, the first engine can be made to run with a higher fuel efficiency while also reducing the fuel consumed by the second engine. Similarly, by running the first engine and the second engine at a higher load and avoiding starting the third engine, the first engine and the second engine can run with a higher fuel efficiency compared to running three engines at a low load.

[0066] As mentioned above, providing Figure 5 As one or more examples. Other examples can be combined with Figure 5 Different than described.

[0067] Figure 6is a graph 600 of one or more exemplary implementations of a hybrid ESS optimization strategy with intelligent adaptive control as described herein. For example, graph 600 may represent a real-time instantaneous calculation of a total load in a power system based on individual loads supported by multiple engines (shown as engines 1-4) and a battery included in the power system, which may be used to implement peak load suppression and load leveling. In some implementations, multiple engines may be associated with certain identifiers, such as a part number or engine serial number that can individually identify or otherwise manage the engines (e.g., engines 1-4 may be associated with corresponding identifiers, such as TKY00274, TKY00275, TKY00276, and TKY00277). Figure 6 In FIG. 1 , the x-axis represents time and the y-axis represents load (eg, in kilowatts).

[0068] like Figure 6 As shown in , during a time period when the total load is below a threshold (e.g., approximately 2000 kW in the example shown), the power controller may initially operate only the first engine (shown as Engine 1 load). Therefore, as shown at reference numeral 610, the power controller may delay starting the second engine until the total load meets the threshold, at which time the second engine may be started to balance the load distributed to the first engine (e.g., to prevent the first engine from having to generate power in excess of the threshold power amount).

[0069] like Figure 6 As further shown in , the first engine and the second engine can be operated until the total load intersects the second threshold. Therefore, as shown in the figure numeral 620, the power controller can discharge the battery (which may have been in a charged state when only the first and / or second engine is running) to handle the peak portion of the load and level or otherwise stabilize the load distributed to the first engine and the second engine. In addition, by discharging the battery to handle the peak portion of the load, the power controller can avoid starting the third engine to save fuel.

[0070] As mentioned above, providing Figure 6 As one or more examples. Other examples can be combined with Figure 6 Different than described.

[0071] Figure 7700 is a graph of one or more exemplary implementations of a hybrid ESS optimization strategy with intelligent adaptive control as described herein. In some implementations, graph 700 illustrates a technique by which the amount of time a powertrain system is driven by a battery and / or an engine may be controlled to provide improved performance, extend the useful life of the battery and / or the engine, etc. For example, repeatedly turning an engine on and off at short intervals may reduce engine performance, may cause premature degradation of engine components, etc. Similarly, charging and discharging a battery for short periods of time may reduce the capacity of the battery due to component degradation or cause other performance issues. Figure 7 In the figures, the x-axis of all four graphs represents time, the y-axis of the first graph represents the total load in the power system (e.g., in kilowatts), the y-axis of the second graph represents the portion of the total load supported by multiple engines in the generator set (e.g., in kilowatts), the y-axis of the third graph represents the number of engines in the generator set that are online at a particular time, and the y-axis of the fourth graph represents the portion of the total load supported by one or more batteries (e.g., in kilowatts).

[0072] Thus, as shown in reference numeral 710, the battery with a high charge level can be discharged initially to handle the total load associated with the power system. As shown in reference numeral 720, the engine can be turned on to handle at least a portion of the load (e.g., based on the battery charge level meeting a threshold), and the engine can remain on for at least a minimum duration (e.g., 20 minutes in graph 700) to ensure that the engine operates at a high efficiency and / or performance level. As shown in reference numeral 730, after the minimum duration has passed and / or the battery has been charged to a level sufficient to handle a larger load, the engine can be turned off, and the engine can remain off for a certain duration (e.g., forty minutes in graph 700) to avoid repeatedly turning the engine on and off. In some cases, keeping the engine on and / or off for at least the minimum duration does not necessarily provide optimal fuel efficiency, but can improve the performance of the engine, extend the service life of the engine, etc.

[0073] As mentioned above, providing Figure 7 As one or more examples. Other examples can be combined with Figure 7 Different than described.

[0074] Figure 8 is a flow chart of an exemplary process 800 for implementing a hybrid ESS optimization strategy with intelligent adaptive control. Figure 8One or more process blocks may be performed by a device, such as a power controller (e.g., power controller 110) and / or another device or group of devices separate from or including the power controller (e.g., an ECM associated with engine(s) 122, a BDP inverter or controller associated with ESS 130, and / or another device associated with embodiment(s) 100, embodiment 200, etc.).

[0075] like Figure 8 As shown in FIG. 8 , process 800 may begin by receiving information related to a load associated with a power system and information related to a set of BSFC maps associated with the power system (block 810). For example, the set of BSFC maps may include one or more BSFC maps that measure the fuel efficiency of one or more engines, one or more battery equivalent BSFC maps based on the cumulative energy cost associated with charging and discharging one or more batteries, the estimated life associated with one or more batteries, the replacement cost of one or more batteries, etc.

[0076] Next, process 800 may include determining a performance indicator using a cost function associated with a plurality of power sources included in the power system (block 820). For example, as described above, the power controller may determine the performance indicator using a cost function associated with a plurality of power sources included in the power system.

[0077] Process 800 may include determining a power distribution among multiple power sources based on the set of BSFC maps and performance indicators to minimize the energy cost associated with the load (block 830). For example, the power controller may generate a dynamic optimization map based on the set of BSFC maps and performance indicators, the dynamic optimization map indicating fuel savings based on different load conditions and charge levels associated with one or more batteries, and the power controller may determine the power distribution among the multiple power sources based on the dynamic optimization map. In another example, the power controller may apply one or more machine learning techniques to determine the power distribution ratio that minimizes the energy cost.

[0078] like Figure 8As further shown in , process 800 may include generating a command to operate the power system based on the power profile, wherein the command indicates a first amount of power associated with operating one or more engines of the plurality of power sources and a second amount of power associated with operating one or more batteries of the plurality of power sources (block 840). For example, the power controller may determine that a machine associated with the power system is in a dynamic positioning zone requiring a backup power source, in which case the command may enable at least two of the plurality of power sources to be operated when the machine is in the dynamic positioning zone. In another example, based on the load failing to meet a first threshold and / or the charge level associated with the one or more batteries failing to meet a second threshold, the command causes the one or more engines to generate a first amount of power to charge the one or more batteries. Additionally or alternatively, based on the load meeting the first threshold and / or the charge level associated with the one or more batteries meeting the second threshold, the command causes the one or more batteries to generate a second amount of power to delay or avoid starting the one or more engines, thereby performing peak load suppression to prevent the first amount of power from exceeding a threshold, providing transient support to the one or more engines based on a response time associated with the load, and the like. In some embodiments, based on health information associated with multiple power sources, the first amount of power can be distributed evenly or unevenly among one or more engines, and the second amount of power can have a positive value to indicate that one or more batteries are to be discharged, or a negative value to indicate that one or more batteries are to be charged.

[0079] Process 800 may include combining Figure 8 Variations of those embodiments described and / or additional embodiments, such as any single implementation or any combination of embodiments described elsewhere herein. Figure 8 An exemplary block diagram of process 800 is shown, but in some examples, Figure 8 Process 800 may include additional blocks, fewer blocks, different blocks, or differently arranged blocks compared to those blocks depicted in . Additionally or alternatively, two or more of the blocks in process 800 may be performed in parallel.

[0080] Industrial Applicability

[0081] The control system disclosed herein can implement a hybrid ESS optimization strategy applicable to any mobile machine having a power system, wherein multiple power sources cooperate to drive the mobile machine and provide power to the main load and / or one or more auxiliary loads under varying conditions. The control system can reduce overall fuel consumption, improve performance, and extend the service life of multiple power sources while also complying with restrictions and / or regulatory requirements by optimizing load sharing under various operating constraints. Therefore, the hybrid ESS optimization strategy implemented by the control system can be used to control a power system having a combination of highly diverse power sources.

[0082] For example, where the power system includes multiple engines, the load can be evenly or unevenly distributed among the multiple engines to ensure that each engine operates at maximum fuel efficiency, reduce the total amount of fuel consumed by each engine, balance or level the instantaneous and / or cumulative loads supported by each engine, maintain and / or extend the service life of each engine, etc. In addition, where the power system includes one or more batteries, the batteries can provide spinning reserve and interruption prevention by acting as a backup component for the generator driven by the engine. The batteries can also be used to implement peak load suppression, to act as a buffer and (individually and / or collectively) level the load supported by the engine, optimize the operating point of the engine and / or battery to reduce maintenance and / or replacement costs, provide instantaneous power to support the generator driven by the engine, etc. Thus, as described herein, the control system can dynamically adapt to different operating conditions and input parameters related to the overall state of the powertrain system to improve performance (e.g., increased acceleration or additional on-demand power), reduce emissions (e.g., locally and / or in a cycle), optimize fuel consumption, reduce total cost of ownership by reducing maintenance and operating costs, improve user comfort (e.g., by achieving silent operation and / or reducing odor and / or safety risks caused by fuel combustion), and the like.

[0083] As used herein, the articles "a" and "an" are intended to include one or more items and may be used interchangeably with "one or more." Additionally, as used herein, the terms "has," "have," "having," and the like are intended to be open-ended terms. Furthermore, the phrase "based on" is intended to mean "based, at least in part, on."

[0084] The above disclosure provides illustration and description, but is not intended to be exhaustive or to limit the embodiments to the precise form disclosed. Modifications and changes may be made in light of the above disclosure or may be acquired from the practice of the embodiments. The description is intended to be considered as exemplary only, and the true scope of the disclosure is defined by the following claims and their equivalents. Even if specific combinations of features are described in the claims and / or disclosed in the description, these combinations are not intended to limit the disclosure of the various embodiments. Although each dependent claim listed below may be directly dependent on only one claim, the disclosure of the various embodiments includes each dependent claim in combination with each other claim in the claim set. .

Claims

1. A method for managing a power system, the method comprising: Received by the device: information relating to the loads associated with the power system, and information relating to a set of brake specific fuel consumption (BSFC) maps associated with the powertrain system; determining, by the device, a performance indicator using a cost function associated with a plurality of power sources included in the power system, the plurality of power sources including one or more engines and one or more batteries; determining, by the device, a power distribution among the plurality of power sources based on the set of BSFC maps and the performance indicator to minimize energy costs associated with the load; as well as generating, by the device, commands for operating the power system based on the power profile, wherein the command indicates a first amount of power associated with operating one or more engines of the plurality of power sources and a second amount of power associated with operating one or more batteries of the plurality of power sources, The set of BSFC maps includes: one or more BSFC maps measuring the fuel efficiency of the one or more engines, and One or more battery equivalent BSFC maps based on cumulative energy costs associated with charging and discharging the one or more batteries, estimated lifespans associated with the one or more batteries, and replacement costs of the one or more batteries.

2. The method according to claim 1, further comprising: generating a dynamic optimization map based on the load associated with the power system, the set of BSFC maps, and the performance indicator, wherein the dynamically optimized map indicates fuel savings based on different load conditions and charge levels associated with the one or more batteries, and The power distribution among the plurality of power sources is determined based on the dynamic optimization map.

3. The method according to any one of claims 1 to 2, further comprising: determining that a machine associated with the power system is in a dynamic positioning zone requiring a backup power source, wherein the command renders at least two of the plurality of power sources operable when the machine is in the dynamic positioning zone.

4. The method according to any one of claims 1 to 3, wherein: The command causes the one or more engines to generate the first amount of power to charge the one or more batteries based on one or more of the loads failing to satisfy a first threshold or a charge level associated with the one or more batteries failing to satisfy a second threshold.

5. The method according to any one of claims 1 to 4, wherein: The command causes the one or more batteries to generate the second amount of power to delay or avoid starting the one or more engines based on one or more of the loads satisfying a first threshold or a charge level associated with the one or more batteries satisfying a second threshold.

6. The method according to any one of claims 1 to 5, wherein: The command causes the one or more batteries to generate the second amount of power to perform one or more of the following operations: perform peak load suppression, which prevents the first amount of power from exceeding a threshold, or provide transient support to the one or more engines based on a response time associated with the load.

7. The method according to any one of claims 1 to 6, wherein: The first amount of power is evenly or unevenly distributed among the one or more engines based on health information associated with the plurality of power sources.

8. The method according to any one of claims 1 to 7, wherein: The second amount of power has a positive value indicating that the one or more batteries are to be discharged or a negative value indicating that the one or more batteries are to be charged.

9. A machine with power system management, comprising: a power system, the power system comprising a plurality of power sources, the plurality of power sources comprising one or more engines and one or more batteries; as well as A power controller comprising one or more processors configured to: receiving information relating to a set of brake specific fuel consumption (BSFC) maps associated with the plurality of power sources; determining a performance indicator using a cost function based on health information associated with the plurality of power sources; as well as generating commands to operate the power system based on a power profile that minimizes energy cost to operate the power system based on information related to the set of BSFC maps, the performance indicator, and a load associated with the power system, The set of BSFC maps includes: one or more BSFC maps measuring the fuel efficiency of the one or more engines, and One or more battery equivalent BSFC maps based on cumulative energy costs associated with charging and discharging the one or more batteries, estimated lifespans associated with the one or more batteries, and replacement costs of the one or more batteries.

Images