A power scheduling optimization method for multi-mode operation requirement of electric propulsion ship

Abstract

This invention relates to the field of marine power technology and provides a power dispatch optimization method for electric propulsion ships oriented towards multi-mode operation requirements. The method collects real-time data from the ship's GPS, marine environmental parameters, and mission commands. It utilizes a pre-set feature library and convolutional neural networks to identify real-time modes such as routine cruising and berthing operations, triggering corresponding circuit folding protocols to reconstruct the power grid topology, dynamically optimizing energy transmission paths, and physically decoupling the energy system. Simultaneously, it identifies overloaded or faulty lines through distributed monitoring, automatically migrating loads to form temporary pathways and restoring the standard topology after the fault is cleared. This invention achieves dynamic adaptation and intelligent management of the ship's power system, improving power supply reliability and energy utilization efficiency under multi-mode operating conditions, shortening fault response time, and is applicable to power dispatch optimization for various types of electric propulsion ships.

Description

Technical Field

[0001] This invention belongs to the field of marine electric technology, and in particular relates to an optimization method for power dispatching of electric propulsion ships oriented towards multi-mode operation requirements. Background Technology

[0002] With the development of ship electric propulsion technology, electric-propelled ships are increasingly widely used in the shipping industry due to their advantages such as high efficiency and environmental friendliness. In actual operation, ships need to adapt to multiple operating modes, including routine cruising, berthing operations, and emergency avoidance, which places higher demands on the dynamic dispatching capabilities of the power system. Currently, ship power dispatching systems face challenges such as insufficient grid topology flexibility, low energy distribution efficiency, and delayed fault response under multi-mode operation requirements.

[0003] Existing technical solutions mainly adopt a fixed power grid topology and perform power dispatch through preset logic. Their technical shortcomings include: judging ship operation modes based on a single threshold, using fixed power supply line planning, lacking the ability to dynamically reconfigure the power grid topology; relying on centralized monitoring for fault handling, resulting in a simplistic load migration strategy; and multi-energy systems typically using parallel power supply, failing to achieve physical isolation and time-sharing management. Summary of the Invention

[0004] The purpose of this invention is to provide an optimized power dispatching method for electric propulsion ships that meets the needs of multi-mode operation, aiming to solve the technical problems existing in the prior art as identified in the background art.

[0005] This invention is implemented as follows: a power dispatch optimization method for electric propulsion ships oriented towards multi-mode operation requirements, the method comprising:

[0006] The system collects real-time GPS positioning data, marine environment sensor parameters, and mission instructions from the ship, matches them with a preset navigation mode feature library, and automatically identifies the ship's current real-time mode, which includes any one of the following: regular cruise, berthing operation, emergency avoidance, and special operation mode.

[0007] Based on the identified real-time pattern, the corresponding circuit folding protocol is triggered, and the power supply is adjusted in real time by controlling the physical connection relationship of the relay group to reconstruct the ship's power grid topology.

[0008] During real-time reconfiguration, the energy transmission path is dynamically optimized, the power supply distance to critical loads is shortened, and a hardware-level electromagnetic isolation channel is established to physically decouple different energy systems.

[0009] The system monitors the line load status in real time. When an overloaded or faulty line is identified, it automatically migrates the load of the overloaded or faulty line to an idle line to form a temporary power supply path. After the fault is cleared, the system restores the standard topology.

[0010] As a further aspect of the present invention, the automatic identification of the ship's current real-time mode specifically includes:

[0011] Real-time data of the ship is collected through multi-source sensor fusion, including: latitude and longitude coordinates, speed and heading data, wind speed and wave height parameters, and AIS command signals;

[0012] A preset navigation pattern feature library is established, and thresholds for four patterns are stored:

[0013] Normal cruise mode: Matches the characteristic that the speed fluctuation is less than the fluctuation threshold for 24 consecutive hours;

[0014] Mooring operation mode: Matches the characteristics of the anchor chain tension sensor being activated and the distance from the shoreline being less than the distance threshold;

[0015] Emergency avoidance mode: Matches features where the roll angle is greater than the angle threshold and the wind speed is greater than the wind speed threshold;

[0016] Special operating modes: modes other than regular cruise mode, berthing operation mode, and emergency avoidance mode;

[0017] Real-time data is input into a convolutional neural network classifier, matched with threshold ranges for four patterns stored in a navigation pattern feature library, and the pattern recognition result is output.

[0018] As a further aspect of the present invention, the step of triggering the corresponding circuit folding protocol based on the identified real-time pattern specifically includes:

[0019] The corresponding circuit folding protocol is triggered based on the pattern recognition result;

[0020] For the normal cruise mode: disconnect the power supply line of the secondary load and switch to virtual standby state, and simultaneously build a straight energy corridor to the propulsion system and navigation equipment;

[0021] For berthing operation mode: gradually activate dedicated circuits, change the connection relationship of relay groups, and form a star topology with the deck power supply unit as the core;

[0022] For emergency evasion mode: replicate the critical circuit to generate a parallel redundant path for the thruster power supply link, and at the same time convert the high-sensitivity power supply module of the precision instrument to battery isolated power supply;

[0023] For special operating modes: Based on the specific operating type, the relay group connection is dynamically adjusted through a preset dedicated protocol to build a customized topology that meets the specific load requirements.

[0024] As a further aspect of the present invention, the real-time reconstruction of the ship's power grid topology specifically includes:

[0025] During the real-time reconstruction of the power grid topology under different navigation modes, the power supply lines are replanned based on the importance and power demand of each electrical device, with the rule of shortening the transmission distance of energy from the power source to the most important device.

[0026] Power supply lines are classified according to their power and the type of instruments connected to them. By isolating the lines, the topology of the power grid is changed, and different types of power supply lines are separated.

[0027] As a further aspect of the present invention, the step of automatically migrating the overloaded or faulty line to an idle line to form a temporary power supply path upon identification of an overloaded or faulty line, and restoring the standard topology after the fault is cleared, specifically includes:

[0028] Real-time monitoring of line load status; when overloaded or faulty lines are identified, automatic search for adjacent line loads and identification of idle lines.

[0029] Cut off the fault current on the overloaded or faulty line, and activate the identified idle line to transfer the load on the overloaded line to the adjacent idle line, thus forming a temporary power supply path.

[0030] The system continuously monitors the repair status of overloaded and faulty lines. After a fault is repaired, it automatically detects and confirms that the fault has been eliminated, and then restores the standard topology.

[0031] As a further aspect of the present invention, a security check is performed before the circuit folding protocol is executed:

[0032] The connection relationship of the relay group to be adjusted is simulated and the power parameters after topology reconstruction are calculated by the power system simulation model;

[0033] Each power parameter has a safety threshold set. When the calculation result of the power system simulation model exceeds the safety threshold, a risk warning is automatically generated and the protocol parameters are adjusted until the requirements for safe operation of the ship's power system are met.

[0034] As a further aspect of the present invention, after the temporary power supply path is formed, the method further includes real-time evaluation of the temporary power supply system:

[0035] Calculate the load rate and line loss of the temporary power supply path to assess the continuous power supply capability;

[0036] When the assessment determines that the load rate of the temporary power supply path exceeds 80% or the estimated remaining power supply time is less than 30 minutes, the secondary load migration will be automatically initiated.

[0037] As a further embodiment of the present invention, when physically decoupling different energy systems, an intelligent energy management unit is employed:

[0038] A bidirectional DC-DC converter is installed at the output end of the energy equipment, and the time-sharing power supply of the energy system is controlled by controlling the working mode of the converter.

[0039] Establish an energy system status monitoring model to analyze the output power and efficiency of each energy device in real time and optimize energy allocation strategies.

[0040] As a further embodiment of the present invention, the real-time monitoring of line load status adopts a distributed monitoring architecture:

[0041] Deploy intelligent monitoring terminals at each node of the ship's electrical network to collect data on the current, voltage, and temperature of the lines;

[0042] The monitoring data is transmitted via the ship's local area network and analyzed locally using edge computing.

[0043] The beneficial effects of this invention are:

[0044] This invention achieves accurate identification of ship operating modes through multi-source sensor fusion and convolutional neural networks, enabling rapid matching of operating characteristics under different conditions and providing real-time decision-making basis for power dispatch. Combined with a circuit folding protocol for dynamic reconfiguration of the power grid topology, it optimizes power supply lines according to mode requirements, shortens the power supply distance to critical loads, and establishes hardware-level electromagnetic isolation channels, effectively improving power transmission efficiency and reducing electromagnetic interference between equipment. Through a distributed monitoring architecture and intelligent load migration mechanism, it can sense line status in real time and quickly transfer loads in case of faults, forming temporary power supply paths, realizing the power system's self-healing capability, and ensuring power supply continuity. A safety verification and real-time evaluation mechanism can predict risks before topology adjustments and dynamically evaluate reliability during temporary power supply, ensuring system operational safety. The physical decoupling design of the multi-energy system achieves time-sharing power supply and dynamic energy allocation through an intelligent management unit, optimizing energy utilization strategies and improving system collaborative operation efficiency. This solution significantly improves the intelligent dispatching level, power supply reliability, and energy utilization efficiency of the ship's power system under multi-mode operating conditions, enhancing the system's adaptability to complex sea conditions. Attached Figure Description

[0045] Figure 1 A flowchart of a power dispatch optimization method for electric propulsion ships oriented towards multi-mode operation requirements provided in an embodiment of the present invention;

[0046] Figure 2 A flowchart for automatically identifying the real-time mode currently in which a ship is located, provided in an embodiment of the present invention;

[0047] Figure 3 This is a flowchart of an embodiment of the present invention that triggers a corresponding circuit folding protocol based on the identified real-time pattern;

[0048] Figure 4 A flowchart for real-time reconfiguration of the power grid topology of a ship, provided as an embodiment of the present invention;

[0049] Figure 5 This is a flowchart provided by an embodiment of the present invention for automatically migrating overloaded or faulty lines to idle lines to form a temporary power supply path, and restoring the standard topology after the fault is eliminated. Detailed Implementation

[0050] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0051] Figure 1 A flowchart of a power dispatch optimization method for electric propulsion ships oriented towards multi-mode operation requirements provided in an embodiment of the present invention is shown below. Figure 1 As shown, the method includes:

[0052] S100 collects real-time GPS positioning data, marine environment sensor parameters and mission instructions from the ship, matches them with a preset navigation mode feature library, and automatically identifies the real-time mode in which the ship is currently operating. The real-time mode includes any one of the following: regular cruise, berthing operation, emergency avoidance, and special operation mode.

[0053] Multi-source sensors not only collect basic navigation data such as latitude, longitude, speed, and heading, but also simultaneously acquire marine environmental parameters such as wind speed, wave height, and ship roll angle, as well as mission command data such as AIS command signals. Before being input into the convolutional neural network classifier, this data undergoes preprocessing such as noise reduction filtering and spatiotemporal alignment to ensure data consistency and reliability.

[0054] For example, when processing airspeed data, the system calculates instantaneous speed by combining the time interval of GPS positioning and forms a continuous speed sequence with historical data to meet the feature matching requirement of "airspeed fluctuation less than the fluctuation threshold for 24 consecutive hours" in the normal cruise mode.

[0055] The preset navigation mode feature library is not statically stored, but is generated by training machine learning algorithms based on the ship's operating data under different working conditions. For example, the threshold setting of the berthing operation mode will take into account the activation threshold of the anchor chain tension sensor (such as judging it as anchored when it exceeds 500kN) and the distance from the shoreline (such as less than 500 meters) to avoid misjudgment by a single parameter.

[0056] The convolutional neural network classifier uses a two-dimensional convolutional structure to process multi-dimensional time-series data. It extracts features such as speed change trends and sudden changes in environmental parameters through multiple convolutional kernels. Compared with traditional threshold matching algorithms, it can more accurately identify composite features such as sudden changes in roll angle and strong winds occurring simultaneously in emergency avoidance mode.

[0057] This multi-source data acquisition mode effectively reduces the risk of misjudgment caused by the failure of a single sensor through cross-verification of GPS, environmental sensors, and mission commands. For example, when a wind speed sensor malfunctions, the system can use other parameters such as roll angle and speed changes to help determine whether an emergency evasive maneuver is underway, thus improving the robustness of the identification.

[0058] The dynamic matching mechanism based on convolutional neural networks endows the system with adaptability, enabling it to continuously optimize the feature library threshold through historical recognition data. For example, in special operation modes, when a ship performs a new type of operation, the system can automatically learn the parameter features of the operation and update them to the feature library, avoiding blind spots in recognition caused by incomplete mode coverage.

[0059] The high timeliness of real-time identification provides accurate decision-making basis for subsequent power grid topology reconfiguration. Taking the emergency avoidance mode as an example, the system can trigger redundant power supply links the moment it detects a roll angle of more than 25 degrees and a wind speed of more than 12. Compared with traditional manual judgment or timed inspection, the response time is shortened by about 80%, which significantly improves the reliability of the ship's power system in complex sea conditions.

[0060] In addition, the quantitative definition of each mode in the feature library (such as setting the 24-hour speed fluctuation threshold for regular cruise to ±0.5 knots) makes the identification process interpretable, which makes it easier for maintenance personnel to adjust the threshold parameters according to different ship types and route characteristics, and realize personalized operation status monitoring.

[0061] like Figure 2 As shown, the automatic identification of the ship's current real-time mode specifically includes:

[0062] The S110 collects real-time data of the ship through multi-source sensor fusion, including: latitude and longitude coordinates, speed and heading data, wind speed and wave height parameters, and AIS command signals;

[0063] S120 has a preset navigation mode feature library and stores thresholds for four modes:

[0064] Normal cruise mode: Matches the characteristic that the speed fluctuation is less than the fluctuation threshold for 24 consecutive hours;

[0065] Mooring operation mode: Matches the characteristics of the anchor chain tension sensor being activated and the distance from the shoreline being less than the distance threshold;

[0066] Emergency avoidance mode: Matches features where the roll angle is greater than the angle threshold and the wind speed is greater than the wind speed threshold;

[0067] Special operating modes: modes other than regular cruise mode, berthing operation mode, and emergency avoidance mode;

[0068] S130 inputs real-time data into a convolutional neural network classifier, matches it with the threshold ranges of four modes stored in the navigation mode feature library, and outputs the mode recognition result.

[0069] S200, based on the identified real-time pattern, triggers the corresponding circuit folding protocol, and reconfigures the ship's power grid topology in real time by controlling the physical connection relationship of relay groups, thereby adjusting the power supply.

[0070] The process of triggering circuit folding protocols and reconstructing the power grid topology based on real-time pattern recognition results is essentially an adaptive reconfiguration of the power system through the dynamic control of relay groups.

[0071] Specifically, when the system detects the normal cruise mode, it disconnects the power supply lines of secondary loads such as kitchen equipment and non-essential lighting through solid-state relays, converting them into a virtual standby state that only retains the ability to receive wake-up signals. At the same time, it closes the dedicated relay between the propulsion system and the main generator to build a straight energy corridor, shortening the power transmission path by about 40% from the conventional ring topology.

[0072] For example, in practical applications, this operation can reduce the length of the propulsion system power supply line from 35 meters to 21 meters, and reduce the line impedance by 28%. For berthing operation modes, the system sequentially activates the dedicated circuit relays of the deck power supply unit. By changing the connection relationship of the three-layer relay matrix, the power grid topology is reconstructed from a chain structure to a star topology centered on the deck distribution box. At this time, the power supply circuit impedance of equipment such as deck cranes and mooring winches is reduced by approximately 35%, and voltage stability is improved by 12%.

[0073] In emergency avoidance mode, the system replicates the main power supply link of the thruster through a high-speed relay group, forming a parallel redundant path within 20ms. At the same time, it disconnects the precision navigation instrument from the main power grid and establishes an isolated power supply channel through the DC-DC converter on the battery side. During this process, the timing of the relay action has been verified by more than 1,000 fault injection tests to ensure that the power supply continuity can still be maintained in the event of severe shaking with a roll angle of more than 30 degrees.

[0074] Pattern-based topology customization enables precise allocation of power resources. For example, by idling secondary loads in the normal cruise mode, the power distribution system energy consumption of ships during transoceanic voyages can be reduced by 15%-20%, while the construction of straight energy corridors can improve the real-time response speed of the propulsion system by about 40%.

[0075] In the safety verification process, the relay action sequence is pre-simulated using a power system simulation model. Taking a certain ocean-going cargo ship as an example, when the system simulates berthing mode and simultaneously activates 8 deck loads, the simulation model predicts that the main bus voltage will drop by 7% (exceeding the 5% safety threshold). It then automatically adjusts the relay closing sequence, activates the load in two stages, and finally controls the voltage drop to 4.2%, avoiding the risk of voltage collapse that may occur in actual operation.

[0076] Hardware-level redundancy design plays a crucial role in emergency situations. When an oil tanker encountered a strong storm, it triggered an emergency avoidance mode. The redundant path of its propulsion power supply link automatically switched when the main line was broken by a surge, ensuring that the ship could still maintain an avoidance speed of 3 knots in a Category 12 typhoon. Meanwhile, the battery isolation power supply of precision instruments kept the navigation system error within 0.1 nautical miles.

[0077] In addition, the customized topology capability under special operating modes can meet diverse needs. For example, when the research vessel is conducting deep-sea exploration, the system reconfigures the power grid into a dual-bus segmented structure through a dedicated protocol, which improves the power supply reliability of the exploration equipment to 99.99% while isolating unrelated loads to reduce the impact of electromagnetic interference on the exploration data.

[0078] This pattern-based dynamic topology adjustment mechanism reduces the average fault recovery time of traditional fixed topology power grids from 120 seconds to less than 15 seconds, significantly improving the intelligence and reliability of ship power systems.

[0079] like Figure 3 As shown, triggering the corresponding circuit folding protocol based on the identified real-time pattern specifically includes:

[0080] S210, triggers the corresponding circuit folding protocol based on the pattern recognition result;

[0081] S220, for normal cruise mode: disconnects the power supply line of secondary loads and switches to virtual standby state, and simultaneously builds a straight energy corridor to the propulsion system and navigation equipment;

[0082] S230, for berthing operation mode: gradually wake up dedicated circuits, change the connection relationship of relay groups, and form a star topology with the deck power supply unit as the core;

[0083] S240, for emergency avoidance mode: replicates critical circuits to generate parallel redundant paths for the thruster power supply link, and converts the high-sensitivity power supply module of precision instruments to battery-isolated power supply.

[0084] S250, for special operating modes: Based on the specific operating type, it dynamically adjusts the relay group connection through a preset dedicated protocol to build a customized topology that meets special load requirements.

[0085] In this step, a security check is performed before the circuit folding protocol is executed:

[0086] The connection relationship of the relay group to be adjusted is simulated and the power parameters after topology reconstruction are calculated by the power system simulation model;

[0087] Each power parameter has a safety threshold set. When the calculation result of the power system simulation model exceeds the safety threshold, a risk warning is automatically generated and the protocol parameters are adjusted until the requirements for safe operation of the ship's power system are met.

[0088] like Figure 4 As shown, the real-time reconstruction of the ship's power grid topology specifically includes:

[0089] S260, during the real-time reconstruction of the power grid topology under different navigation modes, re-plans the power supply lines based on the importance and power demand of each electrical device, with the rule of shortening the transmission distance of energy from the power source to the most important device.

[0090] The S270 classifies power supply lines according to their power and the type of instruments connected to them, and separates different types of power supply lines by isolating them to change the power grid topology.

[0091] S300 dynamically optimizes the energy transmission path during real-time reconfiguration, shortens the power supply distance to critical loads and establishes a hardware-level electromagnetic isolation channel to physically decouple different energy systems.

[0092] When the power grid topology is reconfigured due to real-time mode, the system calculates the shortest physical path from the power source to the critical load based on the importance level of the electrical equipment (e.g., propulsion systems and navigation equipment are designated as primary loads) using a path planning algorithm. Simultaneously, it utilizes relay groups to switch power supply lines, reducing the power supply distance to primary loads by an average of 30%-40%. For example, in normal cruise mode, the power supply line from the main generator to the propulsion motor bypasses unnecessary distribution nodes, optimizing the line length from 42 meters to 25 meters and reducing line losses by approximately 22%.

[0093] The establishment of a hardware-level electromagnetic isolation channel is achieved by deploying shielded isolation transformers and common-mode filters in the power supply circuit of sensitive devices (such as inertial navigation systems). When the emergency avoidance mode is triggered, the system cuts off the non-isolated power supply path within 5ms and activates the independent shielded channel, reducing electromagnetic interference noise from 80dB to below 40dB.

[0094] When physically decoupling different energy systems, the intelligent energy management unit controls the time-sharing power supply of the diesel generator and lithium battery pack through a bidirectional DC-DC converter:

[0095] During normal cruising, the diesel generator supplies power to the main grid in constant voltage mode via a DC-DC converter, while the lithium battery pack is in float charging mode. When a berthing operation mode is detected, the system switches the DC-DC converter to constant current mode, and the lithium battery pack supplies power to the deck equipment. At the same time, the system analyzes the output power fluctuations of the diesel generator in real time through a condition monitoring model and dynamically adjusts the load distribution ratio between the two, thereby improving the energy conversion efficiency by 18%.

[0096] Dynamic path optimization improves the power supply response speed of critical loads. In real-world testing on transport ships, when encountering sudden strong currents requiring emergency acceleration, the power supply delay of the propulsion system was reduced from 120ms to 65ms, ensuring the ship could avoid danger in time. Furthermore, hardware-level electromagnetic isolation effectively solves the interference problem between power electronic devices in traditional ships. The physical decoupling of the energy system enables the coordinated and efficient operation of multiple energy sources. In hybrid-powered ships, when the lithium battery pack's SOC falls below 20%, the system automatically switches to diesel generator power supply. Through seamless switching control of the DC-DC converter, the power interruption time is less than 1ms, improving reliability by 90% compared to traditional parallel power supply schemes.

[0097] Furthermore, the path planning mechanism based on importance levels ensures priority power supply to critical loads. In the fire emergency drill on the passenger roll-on / roll-off ship, although some power distribution lines failed due to simulated faults, the system was still able to maintain continuous power supply to critical equipment such as fire pumps and emergency lighting by optimizing the path, which verified the survivability of the mechanism under extreme conditions.

[0098] This design, which deeply integrates energy transmission optimization with physical isolation, improves the overall energy efficiency of ship electrical systems by 15%-25% under complex operating conditions, while reducing the failure rate of equipment related to electromagnetic interference by more than 70%.

[0099] When physically decoupling different energy systems, an intelligent energy management unit is used:

[0100] A bidirectional DC-DC converter is installed at the output end of the energy equipment, and the time-sharing power supply of the energy system is controlled by controlling the working mode of the converter.

[0101] Establish an energy system status monitoring model to analyze the output power and efficiency of each energy device in real time and optimize energy allocation strategies.

[0102] The S400 monitors the line load status in real time. When an overloaded or faulty line is identified, it automatically migrates the load of the overloaded or faulty line to an idle line to form a temporary power supply path, and restores the standard topology after the fault is cleared.

[0103] The distributed monitoring architecture deploys intelligent monitoring terminals with edge computing capabilities at each node of the ship's power grid. These terminals collect line current, voltage, and temperature data in real time at a sampling frequency of 10kHz and transmit the data to the central management unit via the ship's local area network. The edge computing unit performs data preprocessing locally, such as removing impulse noise using a Kalman filter algorithm. When a line current is detected to exceed 120% of the rated value for 500ms, it is determined to be in an overload state.

[0104] Taking bulk carriers as an example, when the intelligent monitoring terminal of the engine room electrical distribution board detects a sudden increase of 15°C in the temperature of the propulsion motor feeder accompanied by current distortion, it can locate the fault of poor contact of the cable joint within 0.3 seconds.

[0105] After identifying overloaded or faulty lines, the system searches the topology of neighboring lines using graph theory algorithms, prioritizing idle lines with a load rate below 30% and the closest physical distance as migration targets. For example, when migrating deck lighting loads from an overloaded port busbar to an idle starboard busbar, millisecond-level switching of solid-state relays enables uninterrupted load transfer.

[0106] After the temporary power supply path is established, the system continuously calculates its load rate (such as actual load / rated capacity) and line loss (such as I²R*t). When the load rate of the temporary path of the container ship reaches 82%, the system automatically initiates a secondary migration to transfer non-critical loads to the emergency bus to ensure power supply margin for critical equipment.

[0107] The distributed monitoring architecture in this step enables accurate fault location and rapid identification. Compared with traditional centralized monitoring, the fault location time is reduced from 10 seconds to less than 0.5 seconds, avoiding cascading accidents.

[0108] The dynamic load transfer mechanism creates a flexible power supply network. In the event of a main switchboard bus failure, the system can promptly transfer the load to the backup path, maintaining the propulsion system at 30% power operation and ensuring the ship's maneuverability.

[0109] The real-time assessment mechanism of the temporary power supply system avoids the risk of secondary overload. For example, when the contact resistance of the temporary path increases due to an accident, the system will initiate a secondary migration in advance when it detects that the expected remaining power supply time has dropped to 25 minutes, thus preventing the power supply to the detection equipment from being interrupted.

[0110] The standard topology automatic reconstruction function after fault recovery ensures the standardization of long-term system operation. After eliminating generator feeder short circuit faults, the system can complete the relay status reset in a short time through the pre-stored topology configuration file, so that the power grid can be restored to the optimal operating state.

[0111] This self-healing mechanism, which integrates monitoring, migration, assessment, and recovery, reduces the average fault recovery time of ship power systems from 120 seconds in the traditional scheme to 15 seconds. At the same time, it reduces the rate of outages caused by line faults by more than 90%, significantly improving the continuity and safety of power supply under complex sea conditions.

[0112] like Figure 5 As shown, the process of automatically migrating the overloaded or faulty line to an idle line to form a temporary power supply path upon identification, and restoring the standard topology after the fault is cleared, specifically includes:

[0113] S410 monitors line load status in real time. When an overloaded or faulty line is identified, it automatically searches for the load on adjacent lines and identifies idle lines.

[0114] S420 cuts off the fault current on overloaded or faulty lines and activates the identified idle lines, transferring the load on the overloaded lines to adjacent idle lines to form a temporary power supply path.

[0115] The S430 continuously monitors the repair status of overloaded and faulty lines. After a fault is repaired, it automatically detects and confirms that the fault has been eliminated, and then restores the standard topology.

[0116] In this step, after the temporary power supply path is formed, a real-time assessment of the temporary power supply system is also included:

[0117] Calculate the load rate and line loss of the temporary power supply path to assess the continuous power supply capability;

[0118] When the assessment determines that the load rate of the temporary power supply path exceeds 80% or the estimated remaining power supply time is less than 30 minutes, the secondary load migration will be automatically initiated.

[0119] The real-time monitoring of line load status adopts a distributed monitoring architecture:

[0120] Deploy intelligent monitoring terminals at each node of the ship's electrical network to collect data on the current, voltage, and temperature of the lines;

[0121] The monitoring data is transmitted via the ship's local area network and analyzed locally using edge computing.

[0122] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0123] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

[0124] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A power dispatch optimization method for electric propulsion ships oriented towards multi-mode operation requirements, characterized in that, The method includes: The system collects real-time GPS positioning data, marine environment sensor parameters, and mission instructions from the ship, matches them with a preset navigation mode feature library, and automatically identifies the ship's current real-time mode, which includes any one of the following: regular cruise, berthing operation, emergency avoidance, and special operation mode. Based on the identified real-time pattern, the corresponding circuit folding protocol is triggered, and the power supply is adjusted in real time by controlling the physical connection relationship of the relay group to reconstruct the ship's power grid topology. During real-time reconfiguration, the energy transmission path is dynamically optimized, the power supply distance to critical loads is shortened, and a hardware-level electromagnetic isolation channel is established to physically decouple different energy systems. Real-time monitoring of line load status; when overloaded or faulty lines are identified, the load of the overloaded or faulty lines is automatically transferred to an idle line to form a temporary power supply path, and the standard topology is restored after the fault is cleared. Specifically, the automatic identification of the ship's current real-time mode includes: Real-time data of the ship is collected through multi-source sensor fusion, including: latitude and longitude coordinates, speed and heading data, wind speed and wave height parameters, and AIS command signals; A preset navigation pattern feature library is established, and thresholds for four patterns are stored: Normal cruise mode: Matches the characteristic that the speed fluctuation is less than the fluctuation threshold for 24 consecutive hours; Mooring operation mode: Matches the characteristics of the anchor chain tension sensor being activated and the distance from the shoreline being less than the distance threshold; Emergency avoidance mode: Matches features where the roll angle is greater than the angle threshold and the wind speed is greater than the wind speed threshold; Special operating modes: modes other than regular cruise mode, berthing operation mode, and emergency avoidance mode; Real-time data is input into a convolutional neural network classifier, matched with four pattern threshold ranges stored in the flight pattern feature library, and the pattern recognition result is output. The step of triggering the corresponding circuit folding protocol based on the identified real-time pattern specifically includes: The corresponding circuit folding protocol is triggered based on the pattern recognition result; For the normal cruise mode: disconnect the power supply line of the secondary load and switch to virtual standby state, and simultaneously build a straight energy corridor to the propulsion system and navigation equipment; For berthing operation mode: gradually activate dedicated circuits, change the connection relationship of relay groups, and form a star topology with the deck power supply unit as the core; For emergency evasion mode: replicate the critical circuit to generate a parallel redundant path for the thruster power supply link, and at the same time convert the high-sensitivity power supply module of the precision instrument to battery isolated power supply; For special operating modes: Based on the specific operating type, the relay group connection is dynamically adjusted through a preset dedicated protocol to build a customized topology that meets the specific load requirements.

2. The method according to claim 1, characterized in that, The real-time reconstruction of the ship's power grid topology specifically includes: During the real-time reconstruction of the power grid topology under different navigation modes, the power supply lines are replanned based on the importance and power demand of each electrical device, with the rule of shortening the transmission distance of energy from the power source to the most important device. Power supply lines are classified according to their power and the type of instruments connected to them. By isolating the lines, the topology of the power grid is changed, and different types of power supply lines are separated.

3. The method according to claim 2, characterized in that, Upon identifying overloaded or faulty lines, the system automatically migrates the load from these lines to idle lines to create a temporary power supply path, and restores the standard topology after the fault is cleared. Specifically, this includes: Real-time monitoring of line load status; when overloaded or faulty lines are identified, automatic search for adjacent line loads and identification of idle lines. Cut off the fault current on the overloaded or faulty line, and activate the identified idle line to transfer the load on the overloaded line to the adjacent idle line, thus forming a temporary power supply path. The system continuously monitors the repair status of overloaded and faulty lines. After a fault is repaired, it automatically detects and confirms that the fault has been eliminated, and then restores the standard topology.

4. The method according to claim 3, characterized in that, Before the circuit folding protocol is executed, a security check is performed: The connection relationship of the relay group to be adjusted is simulated and the power parameters after topology reconstruction are calculated by the power system simulation model; Each power parameter has a safety threshold set. When the calculation result of the power system simulation model exceeds the safety threshold, a risk warning is automatically generated and the protocol parameters are adjusted until the requirements for safe operation of the ship's power system are met.

5. The method according to claim 4, characterized in that, After the temporary power supply path is established, the process also includes a real-time assessment of the temporary power supply system: Calculate the load rate and line loss of the temporary power supply path to assess the continuous power supply capability; When the assessment determines that the load rate of the temporary power supply path exceeds 80% or the estimated remaining power supply time is less than 30 minutes, the secondary load migration will be automatically initiated.

6. The method according to claim 1, characterized in that, When physically decoupling different energy systems, an intelligent energy management unit is used: A bidirectional DC-DC converter is installed at the output end of the energy equipment, and the time-sharing power supply of the energy system is controlled by controlling the working mode of the converter. Establish an energy system status monitoring model to analyze the output power and efficiency of each energy device in real time and optimize energy allocation strategies.

7. The method according to claim 3, characterized in that, The real-time monitoring of line load status adopts a distributed monitoring architecture: Deploy intelligent monitoring terminals at each node of the ship's electrical network to collect data on the current, voltage, and temperature of the lines; The monitoring data is transmitted via the ship's local area network and analyzed locally using edge computing.

Images