A range extended hybrid marine vessel drive system and method

Abstract

The application provides a range-extending hybrid ship driving system and method, which comprises: a navigation state sensing module for collecting ship navigation state data in real time; a load power prediction module for predicting the ship propulsion load power in a future predetermined time window based on the navigation state data through a pre-trained load power prediction model; a multi-energy collaborative controller for solving an optimal power distribution sequence through a model predictive control algorithm according to the predicted ship propulsion load power, the current power battery state of charge, the current range-extending internal combustion engine operating state and an energy efficiency optimization target function, and generating dynamic power distribution instructions of the power battery and the range-extending internal combustion engine respectively; and a motor driving module for controlling the power battery and the range-extending internal combustion engine to provide electric energy for the ship propulsion motor according to the dynamic power distribution instructions. Through the scheme, the adaptability and stability of the range-extending hybrid ship in various scenes can be improved.

Description

Technical Field

[0001] This invention belongs to the field of marine propulsion system technology, and particularly relates to a range-extended hybrid marine propulsion system and method. Background Technology

[0002] With the rapid development of new energy ship technologies, range-extended hybrid propulsion systems have become an important development direction for ship power systems due to their significant advantages in improving range and energy efficiency. These systems typically combine power batteries and auxiliary power generation units, supplementing pure electric drive with a range extender to meet both zero-emission operation and long-range requirements. Their core lies in multi-energy coordinated configuration and dynamic energy management, involving several key technologies such as power electronic conversion, load power prediction, operating condition identification, and control strategy optimization, which are of great significance to the intelligence and efficiency of ship power systems.

[0003] The key to range-extended hybrid marine propulsion systems lies in achieving efficient coordination and dynamic scheduling among the power battery, internal combustion engine range extender, and other optional energy units. Ideally, the system should be able to adaptively adjust the operating mode and power output of each energy module according to the ship's real-time navigation status, load changes, and environmental conditions to maximize overall energy efficiency and ensure operational stability. However, current technologies have not yet formed a complete solution that balances response speed, control precision, and system robustness.

[0004] Currently, while some solutions achieve flexibility in power replenishment by setting up multiple range extenders and AC / DC converters, their control logic relies excessively on the state of charge of the power battery and lacks a dynamic perception and response mechanism for actual propulsion loads and navigation conditions, resulting in lagging or redundant energy distribution. Another solution introduces new energy sources such as hydrogen fuel cells, which improves energy diversity, but sacrifices system reliability due to the complexity of hydrogen production and supply, and fails to establish a collaborative optimization framework among multiple energy sources, making it difficult to achieve optimal global energy efficiency under varying operating conditions. These shortcomings result in existing systems exhibiting large fluctuations in energy utilization efficiency, slow dynamic response, and rigid control strategies when dealing with complex scenarios such as acceleration, deceleration, berthing, or severe sea conditions, leading to poor adaptability and severely restricting the widespread application of range-extended hybrid vessels in actual shipping. Summary of the Invention

[0005] In view of this, embodiments of the present invention provide a range-extended hybrid ship propulsion system and method to solve the problem of poor adaptability of existing range-extended hybrid ship propulsion systems in complex scenarios.

[0006] In a first aspect of the present invention, a range-extended hybrid marine propulsion system is provided, comprising:

[0007] The navigation status sensing module is used to collect ship navigation status data in real time.

[0008] The load power prediction module is used to predict the ship's propulsion load power within a predetermined time window based on the navigation status data and a pre-trained load power prediction model. The load power prediction model is a bidirectional long short-term memory network model based on an attention mechanism.

[0009] The multi-energy collaborative controller is used to solve the optimal power allocation sequence through model predictive control algorithm based on the predicted ship propulsion load power, the current state of charge of the power battery, the current operating status of the internal combustion engine range extender, and the energy efficiency optimization objective function, and to generate dynamic power allocation commands for the power battery and the internal combustion engine range extender respectively.

[0010] The motor drive module is used to control the power battery and internal combustion engine range extender to provide electrical energy to the ship's propulsion motor according to dynamic power distribution commands.

[0011] In a second aspect of the present invention, a range-extended hybrid ship propulsion method is provided, comprising:

[0012] Real-time collection of ship navigation status data;

[0013] Based on the navigation status data, the ship's propulsion load power within a predetermined time window is predicted by a pre-trained load power prediction model, which is a bidirectional long short-term memory network model based on an attention mechanism.

[0014] Based on the predicted ship propulsion load power, the current state of charge of the power battery, the current operating status of the internal combustion engine range extender, and the energy efficiency optimization objective function, the optimal power allocation sequence is solved by the model predictive control algorithm, and dynamic power allocation commands for the power battery and the internal combustion engine range extender are generated respectively.

[0015] According to the dynamic power distribution command, the power battery and internal combustion engine range extender are controlled to provide electrical energy to the ship's propulsion motor.

[0016] In a third aspect of the present invention, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and executable by the processor, wherein the processor executes the computer program to implement the steps of the method as described in the first aspect of the present invention.

[0017] In a fourth aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the method provided in the first aspect of the present invention.

[0018] In this embodiment of the invention, the ship's propulsion load power within a predetermined time window is predicted by a load power prediction model. By constructing an energy efficiency optimization objective function, a model predictive control algorithm is used to solve for the optimal power allocation sequence and obtain the power allocation command between the power battery and the internal combustion engine range extender. This not only enables forward-looking dynamic perception of the ship's propulsion load, avoiding the problem of energy allocation redundancy or insufficiency caused by relying on lagging battery state information, but also enables the optimal dynamic allocation of power output between the power battery and the internal combustion engine range extender under complex and variable operating conditions. This avoids the bottleneck of rigid control strategies and efficient and stable operation under complex sea conditions in existing solutions. As a result, the adaptability of the range-extended hybrid ship propulsion system in various scenarios can be effectively improved. At the same time, it has good economic benefits, high reliability, and is energy-saving and environmentally friendly. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a schematic diagram of the structure of a range-extended hybrid ship propulsion system according to an embodiment of the present invention;

[0021] Figure 2 Another structural schematic diagram of a range-extended hybrid ship propulsion system provided in one embodiment of the present invention;

[0022] Figure 3 A schematic diagram illustrating the connection relationship between a power battery and an internal combustion engine range extender according to an embodiment of the present invention;

[0023] Figure 4 This is a schematic diagram illustrating the working principle of a multi-energy collaborative controller provided in one embodiment of the present invention;

[0024] Figure 5 A schematic flowchart of a range-extended hybrid ship propulsion method provided in one embodiment of the present invention;

[0025] Figure 6 This is a schematic diagram illustrating the collaborative control effect provided in one embodiment of the present invention;

[0026] Figure 7 This is a schematic diagram of the structure of an electronic device provided in one embodiment of the present invention. Detailed Implementation

[0027] To make the objectives, features, and advantages of this invention more apparent and understandable, the technical solutions of the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the embodiments described below are only some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0028] It should be understood that the terms "comprising" and other similar expressions in the specification, claims, and accompanying drawings of this invention are intended to cover a non-exclusive inclusion, such as a process, method, system, or apparatus that includes a series of steps or units and is not limited to the listed steps or units. Furthermore, "first" and "second" are used to distinguish different objects and are not intended to describe a specific order.

[0029] Please see Figure 1 The present invention provides a schematic diagram of the structure of a range-extended hybrid ship propulsion system, comprising:

[0030] The navigation status sensing module 110 is used to collect ship navigation status data in real time.

[0031] The navigation status perception module 110 is used to collect multi-dimensional physical signals that characterize the ship's navigation conditions in real time, so as to predict the load power required for the ship's navigation based on the current navigation conditions.

[0032] The navigation status perception module includes at least a GPS positioning unit, an inertial measurement unit, a wind speed and direction sensor, a water depth sensor, and a ship attitude sensor.

[0033] The ship navigation status data includes at least the ship's position, speed, heading, three-axis acceleration, three-axis angular velocity, ambient wind speed and direction, water depth, and the ship's roll, pitch and heave attitude angles.

[0034] The GPS positioning unit is used to obtain the ship's real-time geographical location, speed, and heading angle; the inertial measurement unit is used to measure the ship's three-axis acceleration and three-axis angular velocity; the wind speed and direction sensor is used to monitor the environmental wind field vector; the water depth sensor is used to obtain the ship's draft; and the ship attitude sensor is used to monitor the ship's roll angle, pitch angle, and heave displacement.

[0035] The load power prediction module 120 is used to predict the ship's propulsion load power within a predetermined time window based on the navigation status data and a pre-trained load power prediction model. The load power prediction model is a bidirectional long short-term memory network model based on an attention mechanism.

[0036] The input of the load power prediction module 120 is connected to the output of the navigation status perception module 110, and is used to dynamically predict the ship's propulsion load power demand within a specific time window in the future based on the perception data. For example, it can predict the ship's propulsion load power within a time window of 2 to 10 minutes in the future.

[0037] Bidirectional Long Short-Term Memory (BiLSTM) is an improved recurrent neural network consisting of a forward LSTM and a backward LSTM. It enables data prediction by simultaneously considering past and future information of sequence data.

[0038] The load power prediction model's input layer receives a sequence of ship navigation status data containing timestamps. It extracts the forward and backward time dependence features of the sequence data through forward and backward hidden layers, respectively. The hidden states are then weighted and fused by the attention layer. Finally, the fully connected output layer generates point prediction values ​​and confidence intervals for the ship's propulsion load power demand within a predetermined future time window.

[0039] The load power prediction module has a built-in bidirectional long short-term memory network model based on the attention mechanism. Its input layer can receive continuous multi-dimensional navigation state sequence data with timestamps provided by the navigation state perception module, and the fully connected output layer can generate point prediction values ​​and confidence intervals of the ship's propulsion load power demand.

[0040] In some embodiments, the bidirectional long short-term memory network model uses Xavier to initialize weights, and employs a mean squared error loss function and an Adam optimizer (learning rate 0.001) during training; the attention layer uses a scaled dot product attention mechanism with a scaling factor of [missing value]. , The key vector dimension is represented; the model output layer uses a linear activation function, and the confidence interval is calculated based on the prediction error distribution; specifically, during the model training phase, the prediction error sequence on the validation set is recorded, and its mean and standard deviation σ are calculated; during the inference phase, for each prediction point, the upper and lower limits of the confidence interval are calculated using ±1.96σ (corresponding to a 95% confidence level) centered on the point prediction value, where the σ value is updated periodically according to the online prediction error.

[0041] The multi-energy cooperative controller 130 is used to solve the optimal power allocation sequence through model predictive control algorithm based on the predicted ship propulsion load power, the current state of charge of the power battery, the current operating state of the internal combustion engine range extender and the energy efficiency optimization objective function, and to generate dynamic power allocation commands for the power battery and the internal combustion engine range extender respectively.

[0042] The input terminal of the multi-energy coordinating controller 130 is connected to the output terminal of the load power prediction module 120 to receive the predicted load power demand and generate dynamic power allocation instructions for each energy unit based on a preset energy efficiency optimization objective function.

[0043] The multi-energy collaborative controller incorporates an energy efficiency optimization objective function based on a model predictive control framework. This objective function simultaneously considers the system's instantaneous energy efficiency, the rate of degradation of the power battery's health, and the nitrogen oxide emission level of the internal combustion engine range extender.

[0044] Optionally, the mathematical expression of the energy efficiency optimization objective function is as follows:

[0045]

[0046] In the formula, Energy efficiency value, To predict the time-domain step size, For the power battery output power in step k, The state of charge of the power battery at step k is... This is a reference value for the state of charge. The output power of the internal combustion engine range extender in step k is... For internal combustion engine range extenders in terms of power The power generation efficiency (its value is obtained by querying the pre-stored engine universal characteristic MAP diagram, which is constructed based on engine bench test data, stored as a two-dimensional interpolation table, and the efficiency value is calculated using a bilinear interpolation algorithm). The corresponding nitrogen oxide emission function (its value is obtained by fitting an emission characteristic curve, which is based on ISO 8178 standard test data and fitted using a multinomial regression model, the specific equation of which is...) (The coefficients a, b, and c are calibrated using the least squares method), and α, β, γ, and δ are all weighting coefficients. The values ​​of the weighting coefficients α, β, γ, and δ range from 0.001 to 1.0, and the specific values ​​are determined through offline simulation calibration. For example, α=0.0015 is used to balance instantaneous efficiency, β=0.8 is used to balance battery health, γ=0.1 is used to control emissions, and δ=0.05 is used to smooth power fluctuations.

[0047] The optimal power allocation sequence is solved using a model predictive control algorithm, generating power allocation commands for both the power battery and the internal combustion engine range extender. Model predictive control is a control algorithm with model prediction, rolling optimization, and feedback correction as its core mechanisms. By predicting the future state of the system and dynamically adjusting the control strategy, it achieves optimal control of complex systems.

[0048] Optionally, a quadratic programming solver is used to solve the optimal control sequence in the finite time domain in each control cycle based on the current system state and load power prediction, and the first control quantity in the sequence is sent to the actuator, i.e., the motor drive module, as the actual power command.

[0049] For example, the control period is set to 1 second, the quadratic programming solver adopts the interior point method, the obstacle parameter is set to μ=0.1, the maximum number of iterations is 50, and the convergence tolerance is 1e-6; the state constraints are transformed into equality constraints through relaxation variables. For example, the power battery state of charge constraint SOC_min ≤ SOC(k) ≤ SOC_max is transformed into SOC(k) = SOC_ref + s_k, where s_k is a relaxation variable.

[0050] The motor drive module 140 is used to control the power battery and the internal combustion engine range extender to provide electrical energy to the ship's propulsion motor according to the dynamic power distribution command.

[0051] The input terminal of the motor drive module 140 is connected to the output terminal of the multi-energy coordinating controller 130, and is used to receive power distribution commands from the multi-energy coordinating controller 130 and execute corresponding power outputs. The motor drive module may include a power battery and an internal combustion engine range extender to provide electrical energy for the ship's propulsion motor.

[0052] In this embodiment, the load power prediction of the ship's propulsion load is prospectively predicted through a bidirectional long short-term memory network with an attention mechanism. This effectively overcomes the problem of energy allocation redundancy or insufficiency caused by relying on lagging battery state information, improving adaptability in complex navigation scenarios. A multi-energy collaborative optimization framework based on model predictive control is adopted, incorporating instantaneous system energy efficiency, battery health, and emission indicators into a unified objective function for rolling optimization. This enables globally optimal dynamic allocation of power output between the power battery and the internal combustion engine range extender under complex and varying operating conditions, ensuring efficient and stable ship operation. Simultaneously, the various modules of the system are tightly coupled through electrical connections and information interaction interfaces, constructing a complete drive solution that is responsive, precise in control, and robust. This provides solid technical support for the large-scale promotion and application of range-extended hybrid ships in actual shipping, while also being economical and environmentally friendly, effectively reducing energy consumption.

[0053] In some embodiments, the training process of the load power prediction model uses historical navigation data and corresponding actual load power data as samples, and determines the network hyperparameters through time-series cross-validation. During the inference phase, the model receives the latest 50 sets of time-series navigation status data in a sliding window manner and outputs the predicted load power values ​​for a future period.

[0054] In one embodiment, such as Figure 2As shown, the motor drive module 140 includes:

[0055] Power battery 210 is used to provide DC power to the ship's propulsion motor according to dynamic power distribution commands;

[0056] The internal combustion engine range extender 220 is used to start and operate in a preset high-efficiency power range according to the dynamic power distribution command, so as to convert fuel chemical energy into alternating current electrical energy.

[0057] AC / DC converter 230 is used to convert the AC power generated by the internal combustion engine range extender into stable DC power and input it into the power supply bus of the power battery and the marine propulsion motor.

[0058] Marine propulsion motor 240 is used to convert received electrical energy into mechanical energy to drive the propeller.

[0059] The controlled end of the power battery 210 is connected to the first output end of the multi-energy coordinating controller 130, and is used to provide DC power to the ship propulsion motor according to the power distribution command; the controlled end of the internal combustion engine range extender 220 is connected to the second output end of the multi-energy coordinating controller 130, and is used to start and operate in a preset high-efficiency power range according to the power distribution command, converting fuel chemical energy into AC power; the input end of the AC / DC converter 230 is connected to the output end of the internal combustion engine range extender 220, and the output end is connected to the input end of the power battery 210 and the power supply bus of the ship propulsion motor 240 respectively, and is used to convert the AC power generated by the internal combustion engine range extender into stable DC power; the power input end of the ship propulsion motor 240 is connected to the output end of the power battery 210 and the output end of the AC / DC converter 230 respectively, and is used to convert the received electrical energy into mechanical energy to drive the propeller.

[0060] The power battery 210 uses a lithium-ion lithium iron phosphate battery pack. Its management system monitors the voltage, current and temperature of the battery pack in real time, and estimates the state of charge and health of the battery online based on the extended Kalman filter algorithm.

[0061] The internal combustion engine range extender 220 includes a variable speed diesel engine and a permanent magnet synchronous generator. The diesel engine adjusts its fuel injection quantity and intake air quantity through an electronic control unit, so that its operating point is constrained within the high-efficiency range determined by the engine's universal characteristic curve. The three-phase AC power output by the permanent magnet synchronous generator is rectified and filtered by an AC / DC converter unit.

[0062] In one embodiment, such as Figure 3As shown, the power battery is connected to the DC bus via a bidirectional DC / DC converter, which operates in boost mode or buck mode according to the instructions of the multi-energy co-controller; the internal combustion engine range extender is connected to the same DC bus via an AC / DC converter, forming a parallel power supply architecture.

[0063] The power battery is connected to the system DC bus via a bidirectional DC / DC converter, which operates in either boost or buck mode according to the instructions of the multi-energy co-controller.

[0064] The power battery and the internal combustion engine range extender work collaboratively through multi-level interaction. At the electrical level, the power battery is connected to the DC bus via a bidirectional DC / DC converter, while the internal combustion engine range extender is connected to the same DC bus via an AC / DC converter, forming a parallel power supply architecture. At the control level, the multi-energy collaborative controller sends power commands to the power battery unit, with command values ​​ranging from -200 kW to +200 kW, and simultaneously sends power commands to the internal combustion engine range extender unit, with command values ​​ranging from 50 kW to 300 kW. At the condition monitoring level, the power battery unit reports real-time data on voltage, current, temperature, state of charge, and health status, while the internal combustion engine range extender unit reports real-time data on engine speed, output power, fuel consumption rate, and nitrogen oxide emissions, all uploaded at a frequency of 10 Hz.

[0065] In some embodiments, the internal combustion engine range extender is controlled to operate within its high-efficiency power range. Specifically, this is achieved by querying a pre-stored engine universal characteristic MAP (Mapping Map) to map the power command to the corresponding engine speed and torque setpoints. The setpoint tracking is then achieved by adjusting the fuel injection quantity and intake valve opening through the electronic control unit. The MAP is a two-dimensional interpolation table, stored in a CSV file with a resolution of 10 rpm × 10 Nm. The mapping algorithm uses bilinear interpolation to calculate the engine speed n_set and torque T_set. The electronic control unit employs a feedforward-feedback control law. The feedforward term calculates the fuel injection quantity based on the engine dynamic model, while the feedback term uses a PID controller (proportional gain K_p = 0.5, integral time T_i = 0.1 s) to adjust the intake valve opening.

[0066] In one embodiment, under typical navigation conditions of a large inland waterway container ship, the range-extended hybrid vehicle propulsion system begins to execute its integrated control process. The navigation status sensing module can be deployed at key locations on the ship's deck and in the cabins. Its GPS positioning unit continuously acquires the ship's latitude and longitude coordinates, ground speed, and true heading angle data at a sampling frequency of 10 Hz. The inertial measurement unit uses a combination of a three-axis MEMS accelerometer and a fiber optic gyroscope to measure the ship's longitudinal acceleration, lateral acceleration, vertical acceleration, yaw rate, roll rate, and pitch rate in real time, covering an acceleration range of ±2g and an angular velocity range of ±200 degrees / second. Wind speed and direction sensors can be installed on the top of the ship's mast, using ultrasonic wind measurement to monitor ambient wind speed and direction, with a range covering 0 to 60 meters / second. A depth sensor can be installed at the keel of the ship's hull, measuring the ship's draft using high-frequency sonar pulses with an accuracy of ±0.1 meters. The ship's attitude sensors are distributed in the four quadrants of the ship, forward, backward, left, and right. By fusing inertial measurement unit and GPS elevation data, the ship's roll angle, pitch angle, and heave displacement are accurately calculated. The roll angle measurement range is ±30 degrees, the pitch angle measurement range is ±15 degrees, and the heave displacement measurement range is ±2 meters.

[0067] The load power prediction module receives continuous multi-dimensional navigation status sequence data from the navigation status perception module. The data sampling interval is 1 second. Each data packet contains a timestamp, latitude and longitude coordinates, three-dimensional acceleration, three-dimensional angular velocity, wind speed and direction, water depth, and the ship's three-axis attitude angles. The module's built-in attention-based bidirectional long short-term memory network model has a three-layer network structure. The input layer has a dimension of 50 time steps multiplied by 12 feature dimensions. The forward and backward hidden layers each contain 128 memory units. The attention layer uses a scaled dot product attention mechanism to weightedly fuse the forward and backward hidden states. The fully connected output layer contains 3 neurons, which output the point prediction value, upper prediction value, and lower prediction value of the ship's propulsion load power demand within the next 5-minute time window. During model inference, the forward hidden layer processes the input sequence in chronological order, extracting the positive impact features of ship acceleration, turning and other maneuvers on power demand. The backward hidden layer processes the input sequence in reverse chronological order, capturing the inverse dependency of ship deceleration, berthing and other operating conditions on power demand. The attention layer calculates the attention weight of the hidden state at each time step, with the weight value ranging from 0 to 1. Finally, the weighted sum is used to obtain the fused feature vector, which is mapped to the power prediction value through a fully connected layer. The predicted output frequency is 1 Hz.

[0068] like Figure 4As shown, the multi-energy collaborative controller performs optimization calculations with a control cycle of 1 second. Its first input receives the power prediction sequence for the next 5 minutes output by the load power prediction module. The second input receives real-time data on voltage, current, temperature, and state of charge from the power battery unit. The third input receives real-time data on engine speed, output power, and emissions from the internal combustion engine range extender unit. The controller's built-in model predictive control framework-based energy efficiency optimization objective function employs a 30-step prediction time domain. At the beginning of each control cycle, based on the current system state and future load power predictions, it continuously solves for the optimal power allocation sequence within a finite time domain.

[0069] The power battery unit uses a 500 Ah lithium-ion lithium iron phosphate battery pack with a nominal voltage of 600 V and a total energy capacity of 300 kWh. The battery management system monitors the individual cell voltage, total voltage, charging / discharging current, and temperature distribution of the battery pack at a frequency of 100 Hz. The voltage measurement accuracy is ±5 mV, the current measurement accuracy is ±0.5 amperes, and the temperature measurement accuracy is ±0.5 degrees Celsius. A state estimator based on an extended Kalman filter algorithm, using a battery equivalent circuit model, estimates the battery's state of charge (SOC) and state of health online, with an SOC estimation error of less than 2% and a state of health estimation error of less than 3%. The power battery unit is connected to the system's DC bus via a bidirectional DC / DC converter with a rated power of 200 kW and an efficiency greater than 97%. Based on instructions from the multi-energy co-controller, the converter operates in either boost or buck mode. In boost mode, it raises the battery voltage to the 750 V DC bus voltage; in buck mode, it lowers the DC bus voltage to the battery charging voltage.

[0070] The internal combustion engine range extender unit comprises a variable-speed diesel engine and a permanent magnet synchronous generator. The diesel engine has a displacement of 8 liters and a maximum power of 320 kW, while the permanent magnet synchronous generator has a rated power of 300 kW and an efficiency greater than 95%. The diesel engine regulates the fuel injection quantity of its high-pressure common rail injection system and the intake air volume of the turbocharger via an electronic control unit. The injection pressure ranges from 1600 bar to 2200 bar, and the intake valve opening control accuracy is 0.1 degrees. The engine operating point is constrained within the high-efficiency zone determined by the engine's universal characteristic curve. This high-efficiency zone is defined as a specific fuel consumption rate below 210 g / kWh, corresponding to an engine speed range of 1200 rpm to 1800 rpm and a torque range of 800 Nm to 1500 Nm. The three-phase AC power output from the permanent magnet synchronous generator has a frequency range of 40 Hz to 60 Hz and a voltage range of 480 V to 600 V, which is rectified and filtered by an AC / DC converter unit.

[0071] The AC / DC converter unit adopts a three-phase full-bridge controlled rectifier topology with a rated power of 300 kW, a conversion efficiency greater than 98%, and a stable output DC voltage within the range of 750V ± 10V. The converter incorporates an LCL filter to remove high-frequency harmonic components, achieving a total harmonic distortion (THD) of less than 3%. The converter output is directly connected to the input of the power battery unit and the power supply bus of the ship's propulsion motor, forming a unified DC power distribution architecture.

[0072] The ship's propulsion motor is a permanent magnet synchronous motor with a rated power of 400 kW, a peak power of 600 kW, and an efficiency greater than 96%. The motor controller employs a vector control strategy, converting DC power from the battery unit and AC / DC converter unit into three-phase AC power. The frequency control range is between 0 Hz and 200 Hz, and the voltage control range is between 0 V and 750 V. The motor's output mechanical torque is transmitted to the propeller through a reduction gearbox with a reduction ratio of 4.5:1. The propeller has a diameter of 2.5 meters and four blades, achieving a propulsion efficiency of 0.65 at the design speed.

[0073] Figure 5 This is a schematic flowchart of a range-extended hybrid ship propulsion method provided by an embodiment of the present invention. The method includes:

[0074] S501. Real-time collection of ship navigation status data;

[0075] The ship navigation status data includes at least the ship's position, speed, heading, three-axis acceleration, three-axis angular velocity, ambient wind speed and direction, water depth, and the ship's roll, pitch and heave attitude angles.

[0076] S502. Based on the navigation status data, predict the ship's propulsion load power within a predetermined time window using a pre-trained load power prediction model. The load power prediction model is a bidirectional long short-term memory network model based on an attention mechanism.

[0077] The load power prediction model's input layer receives a sequence of ship navigation status data containing timestamps. It extracts the forward and backward time dependence features of the sequence data through forward and backward hidden layers, respectively. The hidden states are then weighted and fused by the attention layer. Finally, the fully connected output layer generates point prediction values ​​and confidence intervals for the ship's propulsion load power demand within a predetermined future time window.

[0078] S503. Based on the predicted ship propulsion load power, the current state of charge of the power battery, the current operating status of the internal combustion engine range extender, and the energy efficiency optimization objective function, the optimal power allocation sequence is solved through the model predictive control algorithm, and dynamic power allocation commands for the power battery and the internal combustion engine range extender are generated respectively.

[0079] Based on the predicted load power demand, the current state of charge of the power battery, the operating status of the internal combustion engine range extender, and the preset energy efficiency optimization objective function, the optimal power allocation sequence is solved through the model predictive control algorithm to generate real-time power commands for the power battery and the internal combustion engine range extender.

[0080] Optionally, the mathematical expression of the energy efficiency optimization objective function is as follows:

[0081]

[0082] In the formula, Energy efficiency value, To predict the time-domain step size, For the power battery output power in step k, The state of charge of the power battery at step k is... This is a reference value for the state of charge. For the output power of the internal combustion engine range extender in step k, For internal combustion engine range extenders in terms of power The power generation efficiency is reduced. Let be the corresponding nitrogen oxide emission function, where α, β, γ, and δ are all weighting coefficients.

[0083] Optionally, a quadratic programming solver is used to solve the optimal control sequence in the finite time domain within each control cycle based on the current system state and load power prediction, and the first control quantity in the sequence is issued to the actuator as the actual power command.

[0084] S504. According to the dynamic power distribution command, control the power battery and internal combustion engine range extender to provide electrical energy to the ship's propulsion motor.

[0085] Optionally, the control of the power battery and internal combustion engine range extender to provide electrical energy to the ship's propulsion motor includes:

[0086] According to the dynamic power distribution command, the power battery is controlled to provide DC power to the ship's propulsion motor;

[0087] According to the dynamic power distribution command, the internal combustion engine range extender is controlled to start and operate in the preset high-efficiency power range to convert fuel chemical energy into alternating current electrical energy.

[0088] The AC power generated by the internal combustion engine range extender is converted into stable DC power by an AC / DC converter and then input into the power supply bus of the power battery and the ship's propulsion motor.

[0089] Ship propulsion motors convert received electrical energy into mechanical energy to drive the propeller.

[0090] Optionally, the power battery is connected to the DC bus via a bidirectional DC / DC converter, which operates in boost mode or buck mode according to the instructions of the multi-energy co-controller; the internal combustion engine range extender is connected to the same DC bus via an AC / DC converter, forming a parallel power supply architecture.

[0091] According to the power command, the power battery outputs DC power to the ship's propulsion motor through a bidirectional DC / DC converter. At the same time, it controls the internal combustion engine range extender to start and operate in its high-efficiency power range. The generated AC power is rectified by the AC / DC converter and then fed into the DC bus. The ship's propulsion motor converts the DC power from the power battery and the AC / DC converter into mechanical torque, driving the propeller to generate thrust, and dynamically adjusts the output power according to real-time navigation status feedback.

[0092] In one embodiment, the specific steps of the range-extended vehicle propulsion method are as follows: Navigation status data is collected in real time via a sensor network deployed on the ship at a frequency of 10 Hz. The collected data packets are packaged and transmitted to the load power prediction module every 100 milliseconds using a CAN bus protocol at a baud rate of 500 kbit / s. The collected navigation status data is input into a pre-trained load power prediction model. During the model inference phase, the latest 50 sets of time-series navigation status data are received in a sliding window manner, with a data window duration of 50 seconds. After a forward calculation process, the predicted load power value for the next 5 minutes and its 90% confidence interval are output, with the prediction value updated at a frequency of 1 Hz. Based on the predicted load power demand, the current state of charge of the power battery, the operating status of the internal combustion engine range extender, and the preset energy efficiency... The objective function is optimized, and the optimal power allocation sequence is solved using a model predictive control algorithm. The control period is set to 1 second, and the optimization calculation time is constrained to within 50 milliseconds. Real-time power commands are generated for the power battery and the internal combustion engine range extender, with a power command resolution of 0.1 kW. Based on the power commands, the power battery outputs DC power to the ship's propulsion motor through a bidirectional DC / DC converter. At the same time, the internal combustion engine range extender is started and operates in its high-efficiency power range. The generated AC power is rectified by the AC / DC converter and fed into the DC bus. The power response time is less than 200 milliseconds. The ship's propulsion motor converts the DC power from the power battery and the AC / DC converter into mechanical torque to drive the propeller to generate thrust. The output power is dynamically adjusted based on real-time navigation status feedback, with a torque control accuracy of ±1 Nm.

[0093] In some embodiments, the training process of the pre-trained load power prediction model uses historical navigation data and corresponding actual load power data for 12 consecutive months as samples, with a total dataset of 2 million time series samples. During training, time series cross-validation is used to determine network hyperparameters, the initial learning rate is 0.001, an exponential decay strategy is adopted, the batch size is 256, and the number of training epochs is 100. The final model achieves a mean absolute percentage error of less than 5% on the test set.

[0094] In some embodiments, the model predictive control algorithm employs an interior-point quadratic programming solver with a maximum number of iterations of 50 and a convergence tolerance of 1e-6. Within each control cycle, based on the current system state and load power prediction, the optimal control sequence within the prediction time domain is solved in a rolling manner for 30 steps. The control time domain is the same as the prediction time domain. The control variables include the output power of the power battery and the output power of the internal combustion engine range extender. The state variables include the state of charge of the power battery and the operating state of the internal combustion engine range extender. The constraints include power battery power limits, upper and lower limits of the state of charge, power limits of the internal combustion engine range extender, and high-efficiency operating constraints.

[0095] The process of controlling the internal combustion engine range extender to operate within its high-efficiency power range is as follows: By querying a pre-stored engine universal characteristic MAP (Mapping Map), the power command is mapped to the corresponding engine speed and torque setpoints. The MAP data resolution is 10 rpm multiplied by 10 Nm of torque. The electronic control unit calculates the required fuel injection quantity and intake valve opening based on the setpoints. The fuel injection quantity control accuracy is 1 mg / cycle, and the intake valve opening control accuracy is 0.1 degrees, achieving a setpoint tracking error of less than 1%.

[0096] Understandable, such as Figure 6 As shown, during ship acceleration, the load power prediction module anticipates the increasing power demand trend 30 seconds in advance. The multi-energy co-controller accordingly increases the discharge power of the power battery and the output power of the internal combustion engine range extender, avoiding the power shortage problem caused by response lag in traditional systems. During ship deceleration, the system anticipates the decrease in power demand, coordinates the recovery of braking energy from the power battery, and simultaneously reduces the output power of the internal combustion engine range extender to maintain its operation in the high-efficiency range. In severe sea conditions, the system dynamically adjusts the power distribution strategy by sensing changes in ship attitude in real time, ensuring the stable operation of the propulsion system.

[0097] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0098] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working process of the system and modules described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0099] Figure 7 This is a schematic diagram of an electronic device according to an embodiment of the present invention. The electronic device is used for ship propulsion control. Figure 7 As shown, the electronic device 7 of this embodiment includes a memory 710, a processor 720, and a system bus 730. The memory 710 includes an executable program 7101 stored thereon. As those skilled in the art will understand, Figure 7 The electronic device structure shown does not constitute a limitation on the electronic device and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0100] The following is combined Figure 7 A detailed introduction to each component of the electronic device:

[0101] The memory 710 can be used to store software programs and modules. The processor 720 executes various functional applications and data processing of the electronic device by running the software programs and modules stored in the memory 710. The memory 710 may mainly include a program storage area and a data storage area. The program storage area may store the operating system, application programs required for at least one function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created according to the use of the electronic device (such as cached data), etc. In addition, the memory 710 may include high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0102] The memory 710 contains an executable program 7101 containing a power calculation method. This executable program 7101 can be divided into one or more modules / units, which are stored in the memory 710 and executed by the processor 720 to implement ship propulsion control, etc. Each module / unit can be a series of computer program instruction segments capable of performing a specific function, describing the execution process of the executable program 7101 in the electronic device 7. For example, the executable program 7101 can be divided into functional modules such as a navigation status perception module, a load power prediction module, a multi-energy coordination module, and a motor drive module.

[0103] The processor 720 is the control center of the electronic device. It connects various parts of the electronic device via various interfaces and lines. By running or executing software programs and / or modules stored in the memory 710, and by calling data stored in the memory 710, it performs various functions and processes data, thereby monitoring the overall status of the electronic device. Optionally, the processor 720 may include one or more processing units; preferably, the processor 720 may integrate an application processor and a modem processor, wherein the application processor mainly handles the operating system, application programs, etc., and the modem processor mainly handles wireless communication. It is understood that the modem processor may not be integrated into the processor 720.

[0104] The system bus 730 is used to connect various functional components within the computer, transmitting data, address, and control information. Its type can be, for example, a PCI bus, an ISA bus, or a CAN bus. Instructions from the processor 720 are transmitted to the memory 710 via the bus, and the memory 710 sends data back to the processor 720. The system bus 730 handles the data and instruction exchange between the processor 720 and the memory 710. Of course, the system bus 730 can also connect to other devices, such as network interfaces and display devices.

[0105] In this embodiment of the invention, the executable program executed by the processor 720 included in the electronic device includes:

[0106] Real-time collection of ship navigation status data;

[0107] Based on the navigation status data, the ship's propulsion load power within a predetermined time window is predicted by a pre-trained load power prediction model, which is a bidirectional long short-term memory network model based on an attention mechanism.

[0108] Based on the predicted ship propulsion load power, the current state of charge of the power battery, the current operating status of the internal combustion engine range extender, and the energy efficiency optimization objective function, the optimal power allocation sequence is solved by the model predictive control algorithm, and dynamic power allocation commands for the power battery and the internal combustion engine range extender are generated respectively.

[0109] According to the dynamic power distribution command, the power battery and internal combustion engine range extender are controlled to provide electrical energy to the ship's propulsion motor.

[0110] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and modules described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0111] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0112] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A range-extended hybrid marine propulsion system, characterized in that, include: The navigation status sensing module is used to collect ship navigation status data in real time. The load power prediction module is used to predict the ship's propulsion load power within a predetermined time window based on the navigation status data and a pre-trained load power prediction model. The load power prediction model is a bidirectional long short-term memory network model based on an attention mechanism. The prediction of ship propulsion load power within a predetermined time window using a pre-trained load power prediction model includes: The input layer of the load power prediction model receives the ship navigation state sequence data containing timestamps. The forward and backward time dependence features of the sequence data are extracted by the forward and backward hidden layers, respectively. The hidden states are weighted and fused by the attention layer. The fully connected output layer generates the point prediction value and confidence interval of the ship propulsion load power demand within a future predetermined time window. The multi-energy collaborative controller is used to solve the optimal power allocation sequence through model predictive control algorithm based on the predicted ship propulsion load power, the current state of charge of the power battery, the current operating status of the internal combustion engine range extender, and the energy efficiency optimization objective function, and to generate dynamic power allocation commands for the power battery and the internal combustion engine range extender respectively. The motor drive module is used to control the power battery and internal combustion engine range extender to provide electrical energy to the ship's propulsion motor according to dynamic power distribution commands.

2. The system according to claim 1, characterized in that, The navigation status perception module includes at least a GPS positioning unit, an inertial measurement unit, a wind speed and direction sensor, a water depth sensor, and a ship attitude sensor. The ship navigation status data includes at least the ship's position, speed, heading, three-axis acceleration, three-axis angular velocity, ambient wind speed and direction, water depth, and the ship's roll, pitch and heave attitude angles.

3. The system according to claim 1, characterized in that, The mathematical expression of the energy efficiency optimization objective function is as follows: In the formula, Energy efficiency value, To predict the time-domain step size, For the power battery output power in step k, The state of charge of the power battery at step k is... This is a reference value for the state of charge. For the output power of the internal combustion engine range extender in step k, For internal combustion engine range extenders in terms of power The power generation efficiency is reduced. Let be the corresponding nitrogen oxide emission function, where α, β, γ, and δ are all weighting coefficients.

4. The system according to claim 1, characterized in that, The process of solving the optimal power allocation sequence using the model predictive control algorithm includes: Using a quadratic programming solver, the optimal control sequence in the finite time domain is solved in each control cycle based on the current system state and the predicted load power. The first control quantity in the sequence is then sent to the motor drive module as the actual power command.

5. The system according to claim 1, characterized in that, The motor drive module includes: The power battery is used to provide DC power to the ship's propulsion motor according to dynamic power distribution commands; An internal combustion engine range extender is used to start and operate in a preset high-efficiency power range according to dynamic power distribution commands, so as to convert fuel chemical energy into alternating current electrical energy. An AC / DC converter is used to convert the AC power generated by the internal combustion engine range extender into stable DC power and input it into the power supply bus of the power battery and the ship's propulsion motor. Marine propulsion motors are used to convert received electrical energy into mechanical energy to drive propellers.

6. The system according to claim 5, characterized in that, The power battery is connected to the DC bus via a bidirectional DC / DC converter, which operates in either boost or buck mode according to the instructions of the multi-energy co-controller; the internal combustion engine range extender is connected to the same DC bus via an AC / DC converter, forming a parallel power supply architecture.

7. A method for driving a range-extended hybrid ship, characterized in that, include: Real-time collection of ship navigation status data; Based on the navigation status data, the ship's propulsion load power within a predetermined time window is predicted by a pre-trained load power prediction model, which is a bidirectional long short-term memory network model based on an attention mechanism. The prediction of ship propulsion load power within a predetermined time window using a pre-trained load power prediction model includes: The input layer of the load power prediction model receives the ship navigation state sequence data containing timestamps. The forward and backward time dependence features of the sequence data are extracted by the forward and backward hidden layers, respectively. The hidden states are weighted and fused by the attention layer. The fully connected output layer generates the point prediction value and confidence interval of the ship propulsion load power demand within a future predetermined time window. Based on the predicted ship propulsion load power, the current state of charge of the power battery, the current operating status of the internal combustion engine range extender, and the energy efficiency optimization objective function, the optimal power allocation sequence is solved by the model predictive control algorithm, and dynamic power allocation commands for the power battery and the internal combustion engine range extender are generated respectively. According to the dynamic power distribution command, the power battery and internal combustion engine range extender are controlled to provide electrical energy to the ship's propulsion motor.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the range-extended hybrid ship propulsion method as described in claim 7.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed, it implements the steps of the range-extended hybrid ship propulsion method as described in claim 7.

Images