Ship hybrid power mode adaptive switching method and system based on working condition recognition

Abstract

The application relates to the technical field of ship hybrid power control, and discloses a ship hybrid power mode adaptive switching method and system based on working condition recognition, wherein the ship hybrid power mode adaptive switching method based on working condition recognition comprises the following steps: step S1, generating synchronous operation data; step S2, extracting working condition parameters; step S3, performing dynamic delay determination; step S4, obtaining a mode switching accommodation amount; step S5, generating a target power mode; step S6, outputting a target torque; and step S7, outputting a mode switching instruction. The application quantitatively evaluates the dynamic torque matching degree in the switching process by constructing a torque establishment path of the candidate power mode and comparing the path with the propeller predicted torque in the time domain; the mode switching accommodation amount is calculated by comprehensively considering the working condition confidence, the torque phase matching degree and the clutch thermal constraint, the target power mode is determined according to the mode switching accommodation amount, and the smoothness of the ship hybrid power mode switching is improved.

Description

Technical Field

[0001] This invention relates to the field of marine hybrid power control technology, and more specifically, to a method and system for adaptive switching of marine hybrid power modes based on operating condition identification. Background Technology

[0002] Hybrid propulsion technology for ships combines the advantages of traditional internal combustion engines and electric propulsion. It can optimize fuel consumption, reduce emissions, and improve maneuverability by adjusting the power source combination under different navigation conditions, and has been gradually promoted and applied in the field of civilian ships in recent years. A typical parallel or series-parallel hybrid power system usually includes components such as a diesel engine, propulsion motor, energy storage battery, friction clutch, and reduction gearbox. The engagement and disengagement of the clutch allows for the switching in and out of different power sources, forming multiple operating modes.

[0003] Existing marine hybrid power systems mostly employ rule-based judgment methods based on preset thresholds for mode switching control. This involves setting switching thresholds based on parameters such as ship speed, propulsion load, or battery state of charge; when a detected parameter crosses the threshold, a corresponding mode switching action is triggered. Some improved methods introduce navigation condition recognition technology, classifying the current navigation state by collecting multi-dimensional data, and then determining the target operating mode based on a preset mapping relationship between operating conditions and modes. The mode switching execution process typically involves the engine governor, motor inverter, and clutch hydraulic system responding to their respective control commands, jointly completing the switching of power sources and torque transfer.

[0004] However, existing mode switching control methods still have several shortcomings in practical applications. First, the operating condition identification data and shaft mechanical measurement data usually come from different sampling channels and time bases, lacking strict time synchronization. This can easily lead to phase deviations between the mode switching decision and the actual mechanical state, affecting the torque continuity during the switching process. Second, the feasibility judgment of mode switching is mostly based on steady-state power or torque parameters, rarely considering the time-domain matching degree of dynamic torque during the switching transition, making it difficult to accurately assess the smoothness of the switching process. Third, clutch thermal load constraints are usually set only as independent protection thresholds, without comprehensive consideration of the operating condition identification results and torque path. This may result in excessively rapid accumulation of clutch slip heat during the switching process, which can affect the service life of the friction pair (i.e., the paired friction system composed of friction plates and mating plates) in the long run. Summary of the Invention

[0005] This invention provides a method and system for adaptive switching of ship hybrid power modes based on operating condition identification, which solves the technical problems mentioned in the background.

[0006] This invention provides an adaptive switching method for ship hybrid power modes based on operating condition identification, comprising the following steps: Step S1: Bind the collected navigation status data and shafting mechanical measurement data to the same timestamp to generate synchronized operation data; Step S2: Utilize synchronized operation data to perform hydrodynamic normalization calculations and feature distance measurements to extract propeller absorbed torque, current identified operating condition, and operating condition confidence level. Step S3: Perform dynamic delay determination and boundary constraint analysis on each candidate power mode together with the propeller absorbed torque and the current identified working condition to generate the propeller predicted torque and candidate establishment torque. Step S4: Perform time-domain integral and friction energy calculation on the propeller predicted torque, candidate establishment torque, operating condition confidence, and clutch slippage state corresponding to each candidate power mode to obtain the mode switching capacity of each candidate power mode. Step S5: Compare the maximum values ​​of each candidate power mode with the corresponding mode switching capacity, filter out the candidate power mode corresponding to the maximum value, and generate the target power mode. Step S6: Substitute the target power mode and the propeller predicted torque into the preset available torque boundary for error allocation and amplitude limiting calculation, and calculate the output motor target torque and engine target torque respectively; Step S7: Encode the target power mode, motor target torque and engine target torque according to the communication protocol based on the synchronization timestamp, and generate a mode switching command output.

[0007] This invention provides a ship hybrid power mode adaptive switching system based on operating condition identification, comprising: The operation data synchronization module is used to bind the collected navigation status data and shaft mechanical measurement data to the same timestamp to generate synchronized operation data; The working condition feature extraction module is used to perform hydrodynamic normalization calculation and feature distance measurement using synchronous operation data, and to extract the propeller absorbed torque, the current identified working condition and the working condition confidence level. The dynamic delay determination module is used to perform dynamic delay determination and boundary constraint analysis on various candidate power modes, together with the propeller absorbed torque and the current identified working condition, to generate propeller predicted torque and candidate establishment torque. The switching capacity calculation module is used to perform deviation time-domain integration and slip energy calculation on the propeller predicted torque, candidate establishment torque, operating condition confidence, and clutch slip state corresponding to each candidate power mode to obtain the mode switching capacity of each candidate power mode. The power mode generation module is used to compare the maximum value of each candidate power mode with the corresponding mode switching capacity, filter out the candidate power mode corresponding to the maximum value, and generate the target power mode. The target torque output module is used to substitute the target power mode and the propeller predicted torque into the preset available torque boundary for error allocation and amplitude limiting calculation, and calculate the output target torque of the motor and the target torque of the engine respectively. The switching command output module is used to encode the target power mode, motor target torque and engine target torque according to the communication protocol based on the synchronization timestamp, and generate the mode switching command output.

[0008] The beneficial effects of this invention are as follows: By synchronously collecting navigation status data and shafting mechanical measurement data and binding them to the same timestamp, this invention can provide input data with a consistent time reference for mode switching decisions, reducing the phase deviation between the operating condition identification result and the mechanical state. By calculating the propeller absorbed torque and identifying the current operating condition and its confidence level, the current navigation status of the ship and its identification reliability can be accurately grasped. By constructing the torque establishment path of candidate power modes and comparing it with the propeller predicted torque in the time domain, the dynamic torque matching degree during the switching process can be quantitatively evaluated. By comprehensively calculating the mode switching capacity based on the operating condition confidence level, torque phase matching degree, and clutch thermal constraints, the multi-dimensional switching influencing factors can be unified into a single quantitative criterion, facilitating objective mode comparison and selection. By determining the target power mode based on the mode switching capacity and allocating the target torque to the engine and motor, better switching smoothness can be achieved while ensuring physical feasibility. By generating synchronously effective mode switching execution commands, the timing consistency of the actions of each actuator can be ensured, further improving the torque continuity during the switching process. This helps to improve the smoothness and reliability of ship hybrid power mode switching and reduce the impact of the switching process on the transmission system. Attached Figure Description

[0009] Figure 1 This is a flowchart of the adaptive switching method for ship hybrid power mode based on operating condition identification according to the present invention; Figure 2 This is a schematic diagram of the ship hybrid power mode adaptive switching system based on operating condition identification according to the present invention; Figure 3 This is a schematic diagram of the overall structure of the marine hybrid propulsion system of the present invention; Figure 4 This is a schematic diagram of the ship operation data acquisition and time synchronization structure of the present invention; Figure 5 This is a schematic diagram of the engine propulsion mode structure of the present invention. Detailed Implementation

[0010] The subject matter described herein will now be discussed with reference to exemplary embodiments. It should be understood that these embodiments are discussed only to enable those skilled in the art to better understand and implement the subject matter described herein, and changes may be made to the function and arrangement of the elements discussed without departing from the scope of this specification. Various processes or components may be omitted, substituted, or added as needed in the examples. Furthermore, features described in some examples may be combined in other examples.

[0011] It should be noted that, unless otherwise defined, the technical or scientific terms used in one or more embodiments of the present invention should have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. The terms "first," "second," and similar terms used in one or more embodiments of the present invention do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" indicate that the element or object preceding the term encompasses the elements or objects listed following the term and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0012] like Figures 1-5 As shown, the adaptive switching method for ship hybrid power modes based on operating condition identification includes the following steps: Step S1: Bind the collected navigation status data and shafting mechanical measurement data to the same timestamp to generate synchronized operation data; Step S2: Utilize synchronized operation data to perform hydrodynamic normalization calculations and feature distance measurements to extract propeller absorbed torque, current identified operating condition, and operating condition confidence level. Step S3: Perform dynamic delay determination and boundary constraint analysis on each candidate power mode together with the propeller absorbed torque and the current identified working condition to generate the propeller predicted torque and candidate establishment torque. Step S4: Perform time-domain integral and friction energy calculation on the propeller predicted torque, candidate establishment torque, operating condition confidence, and clutch slippage state corresponding to each candidate power mode to obtain the mode switching capacity of each candidate power mode. Step S5: Compare the maximum values ​​of each candidate power mode with the corresponding mode switching capacity, filter out the candidate power mode corresponding to the maximum value, and generate the target power mode. Step S6: Substitute the target power mode and the propeller predicted torque into the preset available torque boundary for error allocation and amplitude limiting calculation, and calculate the output motor target torque and engine target torque respectively; Step S7: Encode the target power mode, motor target torque and engine target torque according to the communication protocol based on the synchronization timestamp, and generate a mode switching command output.

[0013] In one embodiment of the present invention, step S1 specifically includes the following steps: Step S101: Collect navigation status data and shafting mechanical measurement data within the same control cycle; Step S102: Perform anti-aliasing filtering on the collected data and convert the engine end torque, motor end torque and clutch end torque to the propulsion shaft side; Step S103: Bind the propeller shaft speed, propeller shaft torque, propeller inlet velocity, pitch status, speed difference between the two ends of the clutch, clutch pressure, engine available torque boundary, motor available torque boundary, and battery available torque status to the same timestamp to generate synchronized operation data.

[0014] Synchronous operation data is a unified data set containing navigation status and shaft mechanical status at the same timestamp.

[0015] The propulsion shaft rotation speed is the rotational speed of the propulsion shaft, which characterizes how fast the propeller rotates.

[0016] The propulsion shaft torque is the actual torque value transmitted by the propulsion shaft, representing the magnitude of the mechanical load transmitted by the propulsion shaft.

[0017] The propeller inflow velocity is the effective water flow velocity entering the propeller disk, which characterizes the speed of the incoming flow to the propeller.

[0018] The pitch state refers to the pitch angle state of the propeller, which represents the angular position of the propeller blades.

[0019] The speed difference between the two ends of the clutch is the speed difference between the driving end and the driven end of the clutch, which characterizes the degree of speed difference between the two ends of the clutch.

[0020] Clutch pressure is the working pressure of the clutch hydraulic system, which characterizes the magnitude of the clutch engagement pressure.

[0021] The engine's available torque boundary is the range of maximum and minimum torque that the engine can output under the current conditions, characterizing the range of the engine's torque output capability.

[0022] The usable torque boundary of a motor is the range of the maximum and minimum torque that the motor can output or absorb under its current state, characterizing the range of the motor's torque output and absorption capabilities.

[0023] The available torque state of a battery is the boundary of the discharge and charge torque that the battery can support in its current state, characterizing the range of torque support capability of the battery.

[0024] Furthermore, all torques from the engine, motor, and clutch must be converted to the propeller shaft side according to the gear ratio. The gear ratio can be obtained from the gearbox nameplate or the ship's propulsion system design documents. The positive direction of torque is uniformly defined as the direction that generates positive thrust; that is, when the torque direction is consistent with the propeller's forward rotation direction and generates forward thrust for the ship, it is considered positive. For example, if a ship's gearbox has a 3:1 gear ratio and the engine output torque is 3000 Nm, then the torque converted to the propeller shaft side is 9000 Nm, and it is considered positive when its direction is consistent with the engine output direction.

[0025] The cutoff frequency of the anti-aliasing filter needs to be determined based on the dynamic characteristics of the ship's propulsion system. The cutoff frequency should be set to at least twice the system's highest frequency of interest, while being less than half the sampling frequency to satisfy the sampling theorem. The main dynamic response frequency of a ship's propulsion system is typically between 0.1 Hz and 10 Hz; therefore, the preferred cutoff frequency for the anti-aliasing filter is 20 Hz, with a range of 10 Hz to 50 Hz. For example, for a system with a control period of 50 milliseconds and a sampling frequency of 20 Hz, the cutoff frequency can be set to 8 Hz to ensure the passage of effective signals while suppressing high-frequency noise.

[0026] The error of the synchronization timestamp should not exceed 10% of the control cycle to ensure that all collected data have sufficient consistency in time. Synchronization can be achieved through the hardware clock triggering mechanism of the controller, with all sensors and control units sharing the same time base signal to complete data sampling and latching at the same time.

[0027] It should be noted that the calculation of the operating condition phase-based switching capacity relies on three types of data: operating condition characteristics, shaft-side torque, and clutch slip state. If the time bases of these data are inconsistent, it will cause a misalignment between the operating condition identification result and the mechanical phase of the propulsion shaft system, making the mode switching capacity unable to accurately reflect the actual capacity of the candidate mode. For example, if the operating condition identification data lags behind the shaft torque data by 20 milliseconds, in scenarios with rapidly changing propeller loads, the identified operating condition may be outdated, resulting in a significant deviation in the calculated capacity. This embodiment uses a strict time synchronization mechanism to ensure that all subsequent calculations are based on the same time base, reducing the distortion in the mode switching capacity calculation caused by data asynchrony and providing a reliable data foundation for subsequently judging the switching feasibility of each candidate mode.

[0028] In one embodiment of the present invention, step S2 specifically includes the following steps: Step S201: Perform hydrodynamic normalization calculation on the propeller inflow velocity, propeller shaft speed and propeller diameter in the synchronous operation data to obtain the propeller advance coefficient; Specifically, the formula for calculating the propeller advance coefficient is as follows: ; in, This indicates the propeller advance coefficient at the moment of the current mode switch determination. This indicates the propeller inflow velocity at the moment the current mode switch is determined. This indicates the propeller revolutions per second calculated from the propeller shaft speed at the moment the current mode switch is determined. Indicates the propeller diameter; The propeller advance coefficient is a dimensionless parameter that characterizes the relative relationship between the propeller inflow velocity and the propeller rotational speed and diameter.

[0029] Step S202: Substitute the propeller advance rate coefficient and pitch state into the torque coefficient graph to obtain the propeller torque coefficient, and calculate the torque by combining the propeller torque coefficient, water density of the navigation area, propeller diameter and propeller shaft speed to obtain the propeller absorbed torque. Specifically, the formula for calculating the torque absorbed by the propeller is as follows: ; in, This indicates the propeller-absorbed torque at the moment of the current mode switch determination. Indicates the water density of the navigation area. This represents the propeller torque coefficient, determined by the propeller advance coefficient and pitch state at the moment of the current mode switch. Indicates the propeller pitch status at the moment of the current mode switch determination; The propeller absorbed torque is the actual load torque absorbed by the propeller from the propulsion shaft and used to propel the water, characterizing the magnitude of the hydrodynamic load applied by the propeller to the propulsion shaft.

[0030] The torque coefficient graph is a set of data provided by the propeller during open water testing or by the manufacturer, characterizing the propeller torque coefficient under different advance rates and pitch conditions.

[0031] Step S203: Normalize the propeller absorbed torque, torque filtering change rate, propeller inflow velocity, bow angular velocity amplitude, and clutch slip amplitude into a working condition identification vector. The operating condition identification vector is a dimensionless feature vector formed by normalizing the propeller torque level, torque change rate, propulsion inflow velocity, bow angular velocity amplitude, and clutch slip amplitude, which represents the comprehensive characteristics of the current ship's navigation and propulsion status.

[0032] Step S204: Perform feature distance measurement between the working condition identification vector and each preset working condition sample matrix to obtain the current identified working condition and working condition confidence. Specifically, the formulas for calculating the feature distance and the current recognition condition are as follows: ; in, This represents the working condition identification vector at the current mode switching determination time up to the [number]th [level]. Feature distance of the preset working condition sample matrix This represents the operating condition identification vector at the moment of current mode switching determination. Indicates the first The test sample center of the preset working condition sample matrix, Indicates the first The covariance matrix of the sample matrix of the preset working conditions. Indicates the current recognition status. This indicates the working condition number corresponding to the minimum feature distance. Specifically, the formula for calculating the confidence level of the operating condition is as follows: ; in, This indicates the confidence level of the operating condition at the time of the current mode switch determination. This represents the minimum feature distance at the moment of the current mode switch determination. This represents the second smallest feature distance at the moment of the current mode switch determination. This represents a numerically stable term.

[0033] Feature distance is a metric that measures how close the current working condition identification vector is to the preset working condition sample distribution, representing the similarity between the current state and the preset working condition.

[0034] The current identified operating condition is the preset operating condition category with the smallest feature distance, which represents the current navigation operating condition of the ship.

[0035] The working condition confidence level is a parameter that characterizes the stability of the working condition identification results and the reliability of the identification results.

[0036] The center of the test sample is the average value of the feature vector under a certain operating condition, which represents the center position of the feature vector under that operating condition.

[0037] The covariance matrix is ​​a matrix that characterizes the fluctuation range and inter-correlation of each characteristic component under a certain operating condition, and is used to calculate the characteristic distance.

[0038] The numerical stability term is a dimensionless value used to avoid zero denominator when distance degenerates.

[0039] Furthermore, each component of the operating condition identification vector needs to be normalized. Each component must be divided by its corresponding rated or maximum value, uniformly mapping each component to the range of 0 to 1. The propeller torque level is divided by the rated torque of the propeller shaft, the torque change rate is divided by the preset maximum torque change rate, the propeller inflow velocity is divided by the inflow velocity corresponding to the ship's design speed, the bow angular velocity amplitude is divided by the maximum turning angular velocity, and the clutch slip amplitude is divided by the rated slip. For example, if the rated torque of the propeller shaft is 10,000 Nm and the current propeller absorbed torque is 5,000 Nm, then the normalized torque level component is 0.5. The typical range of the bow angular velocity amplitude is 0 to 5 degrees per second. During smooth navigation, it is usually below 0.5 degrees per second, and can reach 3 to 5 degrees per second during sharp turns. The typical range of the clutch slip amplitude is 0 to 50 revolutions per minute. The slip is 0 when the clutch is fully engaged, and gradually decreases during engagement.

[0040] The values ​​for propeller diameter and water density in the navigation area need to be determined based on the actual situation. Propeller diameter is an inherent parameter of the ship's propulsion system, determined by the propeller design; specific values ​​can be obtained from propeller drawings or ship design documents. For small and medium-sized civilian vessels, propeller diameter is typically between 2 and 6 meters. Water density in the navigation area is determined based on the type of waterway; 1000 kg / m³ is suitable for freshwater areas, and 1025 kg / m³ for seawater areas.

[0041] The covariance matrix needs to be positive definite to ensure invertibility. When the number of test samples is insufficient or there is a strong correlation between feature components, causing the covariance matrix to be non-invertible, a small regularization term can be added to the diagonal of the matrix to make it a positive definite matrix. The size of the regularization term can be determined according to the number of samples and the feature dimension, and is usually taken as one-hundredth to one-thousandth of the average feature variance.

[0042] The preset operating condition sample matrix must be established through a standardized sea trial procedure. Sea trials should be conducted under various sea states, covering all preset operating conditions, including berthing and unberthing, low-speed maneuvering, acceleration and deceleration, stable cruising, and wave load disturbances. Sufficient sample data should be collected for each operating condition, with the sampling interval consistent with the control cycle. After data collection, outlier removal and smoothing preprocessing are required, followed by calculation of the sample center and covariance matrix for each operating condition category. The sample selection criteria are complete data within a stable operating condition range, excluding transitional and abnormal fluctuation data. For example, for stable cruising, data should be collected for at least 5 minutes after the ship's speed stabilizes, from which at least 100 valid samples should be selected.

[0043] The value of the numerical stability term needs to balance numerical stability and calculation accuracy. If the value of this parameter is too small, it may not completely avoid the case where the denominator is zero, while if the value is too large, it may affect the calculation accuracy of the operating condition confidence level. A value range of 0.000001 to 0.0001 can meet the above requirements, with a preferred value of 0.00001.

[0044] It should be noted that the propeller-absorbed torque is the foundation for subsequent torque prediction paths. The current identified operating condition determines the candidate mode set and parameters, while the operating condition confidence level is used to calculate the subsequent mode switching capacity. These three factors together constitute the input basis for subsequent mode switching decisions. For example, when the ship is near the boundary between acceleration and cruising operating conditions, the characteristic distances of the two conditions may be relatively close. In this case, the operating condition confidence level is low, and the mode switching capacity will decrease accordingly. This embodiment accurately grasps the ship's current navigation state and its identification reliability by calculating the propeller-absorbed torque and identifying the current operating condition and its confidence level, providing reliable input data for subsequently constructing candidate mode torque paths and calculating the mode switching capacity.

[0045] In one embodiment of the present invention, step S3 specifically includes the following steps: Step S301: Read the switching duration, mode lag time, and torque of each candidate power mode according to the current identification working condition to establish the rate boundary; The switching duration is the total time required to complete the switching of a candidate power mode, representing the length of time the mode switching process takes.

[0046] Mode lag time is the equivalent delay time between the issuance of the control command and the actual start of change in the shaft-side torque, characterizing the response lag characteristics of the actuator.

[0047] The torque build-up rate boundary is the maximum speed at which a power source or actuator can increase or decrease shaft-side torque per unit time, characterizing the upper limit of the actuator's torque variation capability.

[0048] The candidate power mode is a hybrid propulsion mode that can be selected, representing the range of options for mode switching.

[0049] Step S302: The propeller absorbed torque and torque filtering change rate are predicted in the time domain to generate the propeller predicted torque. Specifically, the formula for calculating the propeller's predicted torque is as follows: ; in, Indicates the first The time offset of the candidate dynamic mode after the current mode switching determination time. The predicted torque of the propeller at that location, This indicates the propeller-absorbed torque at the moment of the current mode switch determination. This represents the rate of change of torque filtering obtained through causal filtering at the current mode switching determination time. Indicates the first The duration of the switching of candidate power modes; The propeller predicted torque is the propeller torque demand path during the switching process, predicted based on the current propeller absorbed torque and torque change rate, representing the propeller load change trend in the short term.

[0050] Step S303: Starting with the propulsion shaft torque, perform dynamic delay determination and boundary constraint analysis on each candidate power mode to generate candidate establishment torque; Specifically, the formula for calculating the candidate establishment torque is as follows: ; in, Indicates the first The time offset of the candidate dynamic mode after the current mode switching determination time. Candidate torque at the location, This indicates the propeller shaft torque at the moment the current mode switch is determined. A sign function representing the difference between the predicted propeller torque endpoint and the propulsion shaft torque. Indicates the first The torque establishment rate boundary for each candidate power mode. Indicates the first The model lag time of the candidate dynamics modes, Indicates the first Candidate dynamic modes in time offset The positive part of the function at; The candidate power modes include pure electric propulsion mode, engine propulsion mode, hybrid boost mode and engine charging mode, and also include the current operating mode.

[0051] The candidate setup torque is the actual axle-side torque path that a candidate mode can establish under execution lag and torque ramp-up limitations, characterizing the actual torque change that the candidate mode can provide during the switching process.

[0052] Furthermore, the mode switching parameter table needs to be established through shipboard commissioning and type testing. During the shipboard commissioning phase, multiple tests should be conducted on each mode switching process under mooring and sea trial conditions, recording the entire process data from command issuance to torque build-up completion. Type testing verifies the response characteristics of each mode under more stringent test conditions; test conditions should cover different operating conditions such as speeds, torque levels, and temperatures. When processing data, the average value of multiple tests is taken as the nominal parameter, while retaining a certain safety margin. For example, if the test results show that the average switching lag time from pure electric propulsion to engine propulsion mode is 300 milliseconds and the average torque build-up rate is 500 Nm / s, then 350 milliseconds and 450 Nm / s can be taken as nominal values ​​in the parameter table to ensure reliability.

[0053] The torque filtering rate of change needs to be implemented using causal filtering; the filter should only use current and historical data for calculation, excluding future data, to ensure the feasibility of online calculation by the real-time controller; causal filtering methods such as first-order inertial filtering or moving average filtering can be used; the filtering time constant needs to be determined based on the actual variation characteristics of the propeller torque, filtering out high-frequency noise while ensuring response speed. For example, using moving average filtering with a sliding window length of 10 sampling points, for a 50-millisecond control cycle, the corresponding time window is 500 milliseconds, which can effectively smooth noise and track torque change trends in a timely manner.

[0054] The positive part function represents the larger of the input value and zero; that is, it outputs the input value when the input is positive, and outputs zero when the input is negative or zero. In the torque build-up path, the positive part function is used to reflect the physical characteristic that the torque does not begin to climb before the lag time. For example, if the mode lag time is 300 milliseconds, then in the first 300 milliseconds after the switch starts, the result of time offset minus lag time is negative, and after processing by the positive part function, the output is zero, and the torque does not begin to climb; after 300 milliseconds, the result is positive, and the torque begins to climb according to the rate boundary.

[0055] It is important to note that the short-term propeller torque demand and the actual shaft torque that the candidate modes can establish are two independent dynamic processes. Traditional mode switching methods only consider steady-state power matching and do not quantitatively analyze the dynamic torque path during the switching process, which is a crucial prerequisite for evaluating mode switching smoothness. For example, in scenarios where the propeller load increases rapidly, although the steady-state torque capability of a candidate mode may be sufficient to meet the demand, its torque build-up rate may be slow and significantly delayed. In the first half of the switching process, it may not be able to provide sufficient torque in time, leading to a decrease in propeller shaft speed. This embodiment aligns the two dynamic processes on the same time axis, enabling accurate prediction of the propeller's dynamic torque demand and the actual torque build-up capability of each candidate mode during the switching process. This provides data support for subsequent calculations of torque phase misalignment and evaluation of mode switching smoothness.

[0056] In one embodiment of the present invention, step S4 specifically includes the following steps: Step S401: Integrate the absolute deviation between the propeller predicted torque and the candidate established torque over the same switching duration, and normalize it using the propulsion shaft rated torque and switching duration to obtain the torque phase misalignment. Specifically, the formula for calculating the torque phase misalignment is as follows: ; in, Indicates the first The torque phase misalignment of the candidate power mode at the moment of current mode switching determination. Indicates the rated torque of the propulsion shaft. Indicates the first The duration of the switching of candidate power modes, Indicates the first The time offset of the candidate dynamic mode after the current mode switching determination time. The predicted torque of the propeller at that location, Indicates the first The time offset of the candidate dynamic mode after the current mode switching determination time. Candidate torque at the location; The torque phase misalignment is the normalized absolute deviation integral between the propeller predicted torque and the candidate established torque over the switching duration, representing the degree of temporal phase matching between the two.

[0057] The rated torque of the propulsion shaft is the maximum continuous operating torque allowed by the propulsion shaft design. It is used to normalize torque deviation and characterizes the torque carrying capacity of the propulsion shaft.

[0058] Step S402: Integrate the absolute value of the product of the clutch slip torque and the clutch slip angular velocity corresponding to each candidate power mode, and normalize it using the preset heat input upper limit to obtain the slip heat input ratio. Specifically, the formula for calculating the sliding friction heat input ratio is as follows: ; in, Indicates the first The ratio of sliding heat input to the candidate power mode at the moment of current mode switching determination. Indicates the preset upper limit of hot input. Indicates the first The time offset of the candidate dynamic mode after the current mode switching determination time. Clutch slip torque at the point, Indicates the first The time offset of the candidate dynamic mode after the current mode switching determination time. Clutch slip angular velocity at the point; The slip friction heat input ratio is the ratio of the actual slip friction heat input of the clutch during the switching process to the maximum heat input allowed for a single switching, which characterizes the relative level of the clutch's thermal load.

[0059] The preset upper limit of heat input is the maximum heat input of the clutch friction pair allowed in a single mode switch, which characterizes the thermal load bearing capacity of the clutch in a single switch.

[0060] Step S403: Perform exponential decay calculation on the working condition confidence, torque phase misalignment and sliding friction heat input ratio to obtain the mode switching capacity of each candidate power mode; Specifically, the formula for calculating the capacity for mode switching is as follows: ; in, Indicates the first The mode switching capacity of each candidate power mode at the current mode switching determination time. This indicates the confidence level of the operating condition at the time of the current mode switch determination. Indicates the first The torque phase misalignment of the candidate power mode at the moment of current mode switching determination. Indicates the first The ratio of sliding heat input to the candidate power mode at the moment of current mode switching determination.

[0061] The mode switching capacity is a dimensionless comprehensive parameter that characterizes the feasibility of a candidate mode to absorb the propeller torque under the current operating conditions and mechanical phase.

[0062] Furthermore, the clutch slip torque and slip angular velocity need to be predicted based on the current state and the switching process. The slip torque can be calculated based on the clutch pressure and friction coefficient. The change in clutch pressure over time is determined by the response characteristics of the clutch hydraulic system. The slip angular velocity is the difference between the angular velocity of the clutch driving end and the angular velocity of the driven end. The speed of the driving end is determined by the engine or motor side, and the speed of the driven end is determined by the speed of the drive shaft. The prediction needs to be combined with the switching process of the target mode, and the slip torque and slip angular velocity at each moment need to be calculated separately. For example, during clutch engagement, the pressure increases from the initial value at a certain slope, and the slip torque increases accordingly. The slip angular velocity gradually decreases synchronously with the speed of both ends until the slip is zero after full engagement.

[0063] Numerical integration can be performed using the trapezoidal integration method or the rectangular integration method. The accuracy requirements are met as long as the integration step size matches the system control cycle. The principle for selecting the integration step size is to minimize the computational load while ensuring calculation accuracy. For the mode switching process of a ship propulsion system, the control cycle is typically tens of milliseconds. For example, if the switching duration is 2 seconds and the control cycle is 50 milliseconds, then the number of integration steps is 40. The trapezoidal integration method is used to meet the engineering application requirements, which will not be elaborated further here.

[0064] The exponential decay calculation indicates that the larger the torque phase misalignment and the ratio of frictional heat input, the more exponentially the switching smoothness decreases. This non-linear relationship better reflects the actual characteristics of switching smoothness: when the misalignment and heat input are small, the switching smoothness decreases slowly and remains within an acceptable range; however, when the misalignment and heat input exceed a certain threshold, the switching smoothness decreases rapidly, and the switching smoothness deteriorates significantly. For example, when the sum of the torque phase misalignment and the ratio of frictional heat input is 0.1, the switching smoothness is approximately 0.9, which is still at a relatively good level; when the sum is 1.0, the switching smoothness is approximately 0.37, and the switching smoothness has clearly decreased.

[0065] It should be noted that operating condition identification stability, shaft torque phase matching degree, and clutch thermal constraint are three independent influencing factors. In traditional methods, these factors are usually set with thresholds as independent constraints. When multiple thresholds conflict, it is difficult to make a unified decision. For example, traditional methods may set torque deviation thresholds and slip friction thermal thresholds separately. When the two thresholds conflict, it is difficult to determine the comprehensive feasibility of mode switching. This embodiment integrates the three influencing factors into a single mode switching tolerance, realizing a comprehensive quantitative evaluation of the smoothness of candidate mode switching. This transforms mode switching selection from an empirical threshold judgment to a decision based on objective calculation, improving the accuracy and reliability of mode switching.

[0066] In one embodiment of the present invention, step S5 specifically includes the following steps: Step S501: Determine the candidate mode set according to the current identification condition, battery available torque state, engine available torque boundary, motor available torque boundary and clutch engagement available state; The candidate pattern set is the set of candidate dynamic patterns that are allowed to participate in the comparison under the current identification conditions, representing the range of patterns that can be selected under the current conditions.

[0067] Step S502: For candidate power modes whose torque envelope cannot cover the predicted torque endpoint value of the propeller, set the corresponding mode switching capacity to a constant of 0. The physical torque envelope is the range of achievable torques jointly defined by the engine, motor, battery, clutch, and transmission system under the current conditions, representing the actual torque boundary that the system can provide.

[0068] Step S503: Compare the maximum values ​​of the mode switching capacity corresponding to each candidate power mode in the candidate mode set to obtain the candidate power mode corresponding to the maximum value. Specifically, the calculation formula for the target dynamic mode is as follows: ; in, This indicates the target power mode at the time of the current mode switching determination. Indicates the current recognition status The corresponding set of candidate patterns Indicates the first The mode switching capacity of each candidate power mode at the current mode switching determination time. This indicates selecting the candidate power mode corresponding to the maximum value of mode switching capacity within the candidate mode set; If at least two candidate power modes have the same mode switching capacity within a preset numerical accuracy, then the target power mode is generated according to the preset mode sequence.

[0069] The preset numerical precision is the minimum difference between two values ​​that the controller can distinguish, used to determine whether two load quantities are numerically equal.

[0070] The preset mode sequence is a priority sequence of modes fixed at the time of loading, which is used to solve the selection problem when multiple candidate modes have the same acceptance volume.

[0071] Furthermore, the candidate mode set for berthing and unberthing conditions includes pure electric propulsion mode and current operating mode to ensure flexibility in low-speed maneuvering and low emission requirements; the candidate mode set for low-speed maneuvering conditions includes pure electric propulsion mode, engine propulsion mode, and current operating mode; the candidate mode set for acceleration and deceleration conditions includes all four modes to meet rapid response requirements; the candidate mode set for stable cruise conditions includes engine propulsion mode, hybrid boost mode, engine charging mode, and current operating mode to prioritize economy; and the candidate mode set for wave load disturbance conditions includes hybrid boost mode, engine propulsion mode, and current operating mode to ensure sufficient power reserves.

[0072] The engine torque boundary is determined by the engine's current speed, temperature, and operating status; the electric motor torque boundary is determined by the electric motor's current speed, temperature, and battery status; the battery torque boundary is calculated from the available charging and discharging power output by the battery management system; and the clutch torque boundary is determined by the clutch's current pressure and friction coefficient. The physical torque envelope is the intersection of the torque boundaries of all these components, representing the actual torque range achievable by the system. For example, if the engine's current maximum output torque is 8000 Nm, the electric motor's current maximum output torque is 3000 Nm, and the battery's current maximum discharge power corresponds to a torque of 2000 Nm, then the upper limit of the physical torque envelope in hybrid boost mode is 10000 Nm, limited by the battery's discharge capacity.

[0073] The specific value of the preset numerical precision is determined by the controller hardware; this parameter depends on the controller's word length and numerical representation method, and is usually the smallest unit that the controller can represent. For a 32-bit floating-point controller, the preset numerical precision can be 0.000001; for a 16-bit fixed-point controller, the preset numerical precision can be the minimum resolution corresponding to the word length.

[0074] The priority setting of the preset mode sequence should follow the principle of stability first; the current operating mode should be placed at the top of the sequence to ensure that the current mode is maintained first when the load capacity is equal, reducing unnecessary mode switching; the priority of other modes can be determined according to the system's operational stability and switching frequency requirements, with modes having lower switching impact having higher priority. For example, if a ship's preset mode sequence from high to low is: current operating mode, hybrid booster mode, engine propulsion mode, engine charging mode, and pure electric propulsion mode, then when the load capacity of two modes is equal, the mode that appears earlier in the sequence should be selected.

[0075] It should be noted that mode switching selection needs to consider both physical feasibility and smoothness. A physically unfeasible mode, even with excellent smoothness, cannot be the target mode. Among physically feasible modes, the one with the best smoothness should be selected. For example, although a candidate mode may have a high calculated load capacity, if its required torque exceeds the current battery's discharge capacity, this mode cannot be selected as the target mode, otherwise, the switching will fail. This embodiment first filters out physically unfeasible candidate modes using physical torque envelope screening, then selects the target mode with the best smoothness based on the mode switching load capacity. Simultaneously, a preset mode sequence resolves ambiguity when values ​​are equal, ensuring that the selected target power mode is physically feasible and has a high mode switching load capacity.

[0076] In one embodiment of the present invention, step S6 specifically includes the following steps: Step S601: Read the motor participation coefficient and engine participation coefficient according to the target power mode, and determine the motor available torque boundary based on the battery available torque status; The motor participation coefficient is a binary function that characterizes whether the motor participates in the shaft-side torque output in the target mode.

[0077] The engine participation coefficient is a binary function that characterizes whether the engine participates in the axle-side torque output under the target mode.

[0078] The available torque boundary of the motor is the upper and lower limits of the motor shaft torque determined by the battery's available torque state and the motor's thermal limit, representing the range of torque that the motor can currently safely output or absorb.

[0079] Step S602: Substitute the difference between the predicted propeller torque endpoint value and the engine efficiency torque corrected by the engine participation coefficient into the available torque boundary of the motor for amplitude limiting calculation to obtain the target torque of the motor. Specifically, the formula for calculating the target torque of the motor is as follows: ; in, Indicates the target torque of the motor. This represents the motor participation factor in the target power mode. The motor participation factor is a constant of 1 when the motor participates in the shaft-side torque output and a constant of 0 when the motor does not participate in the shaft-side torque output. This represents the engine participation coefficient in the target power mode. The engine participation coefficient is a constant of 1 when the engine participates in the axle-side torque output and a constant of 0 when the engine does not participate in the axle-side torque output. This indicates the predicted propeller torque at the end of the switching duration corresponding to the target power mode. Indicates the switching duration corresponding to the target power mode. This represents the engine efficiency torque obtained by interpolation from the engine efficiency map at the current propulsion shaft angular velocity. This indicates the propulsion axis angular velocity at the moment of the current mode switch determination. This represents the lower bound of the available motor torque, determined by the battery's available torque state and the target power mode. This represents the upper bound of the available motor torque, determined by the battery's available torque state and the target power mode. Represents the amplitude limiting function; Engine efficiency torque is the torque value obtained by interpolation of the engine's high-efficiency operating range map at the current propulsion shaft angular velocity, representing the torque corresponding to the engine's higher efficiency operating point at the current speed.

[0080] Step S603: Substitute the difference between the predicted propeller torque endpoint value and the motor target torque into the engine available torque boundary for amplitude limiting calculation to obtain the engine target torque; Specifically, the formula for calculating the engine target torque is as follows: ; in, Indicates the engine's target torque. This represents the engine participation coefficient under the target power mode. This indicates the predicted propeller torque at the end of the switching duration corresponding to the target power mode. Indicates the target torque of the motor. This indicates the lower limit of the engine's available torque in the target power mode. This indicates the upper limit of the engine's available torque in the target power mode. Represents the amplitude limiting function; Among them, the target torque of the electric motor and the target torque of the engine are both propulsion shaft side torques.

[0081] The engine's available torque boundary is the upper and lower limits of the shaft-side torque that the engine can safely output in target mode, representing the torque range in which the engine can currently operate stably.

[0082] Furthermore, the engine's high-efficiency operating range map is derived from engine bench tests or data provided by the manufacturer; this map describes the engine's efficiency level at different speeds and torques, typically represented by contour lines. Interpolation methods can employ two-dimensional linear interpolation or spline interpolation to calculate the corresponding torque value based on the current speed and target efficiency level. The high-efficiency operating range is generally defined as the region where efficiency exceeds 90% of the maximum efficiency. For example, if an engine's maximum efficiency point at 1500 rpm corresponds to a torque of 6000 Nm and an efficiency of 42%, then the torque range of the high-efficiency operating range is approximately 4500 Nm to 7500 Nm, with an efficiency not lower than 37.8% within this range.

[0083] The available torque boundaries for both the motor and the engine need to be updated in real time. The update frequency is consistent with the control cycle, with the latest boundary values ​​read in each control cycle. The motor torque boundary is affected by factors such as battery status, motor temperature, and inverter temperature, and is calculated and output in real time by the motor control system. The engine torque boundary is affected by factors such as engine speed, temperature, boost pressure, and fuel system status, and is calculated and output in real time by the engine control system. Real-time updates ensure that torque distribution is always based on the current actual capacity, avoiding torque commands becoming unexecutable due to boundary changes. For example, when battery temperature rises, causing a decrease in discharge capacity, the upper limit of the available motor torque will decrease accordingly, and torque distribution will automatically adjust to adapt to the new boundary.

[0084] The function of a limiting function is to restrict the input value between a lower and upper bound. When the input value is greater than the upper bound, the upper bound value is output; when the input value is less than the lower bound, the lower bound value is output; and when the input value is between the upper and lower bounds, the input value itself is output. In torque calculation, the limiting function is used to ensure that the target torque does not exceed the capability range of each power source. For example, if the usable torque range of a motor is -2000 Nm to 3000 Nm, and the calculated target torque of the motor is 3500 Nm, then after limiting, 3000 Nm will be output to ensure that the motor can execute the torque command.

[0085] It should be noted that after determining the mechanically feasible target mode, the torque distribution strategy needs to simultaneously consider the continuity of propulsion torque and the system's fuel economy. The distribution method of using an engine high-efficiency benchmark plus motor residual compensation—that is, first allowing the engine to operate in its high-efficiency region, then having the motor preferentially absorb the difference between the target torque and the engine's high-efficiency torque, and finally having the engine handle the remaining portion—can effectively balance these two objectives. For example, under stable cruising conditions, if the target propulsion torque is 7000 Nm and the engine's high-efficiency torque is 6000 Nm, then the motor outputs 1000 Nm to compensate for the difference, ensuring the engine always operates at its high-efficiency point. This embodiment, through this torque distribution strategy, achieves a reasonable torque distribution between the engine and the motor, improving the fuel economy of the marine hybrid power system while ensuring continuous and stable propulsion torque.

[0086] In one embodiment of the present invention, step S7 specifically includes the following steps: Step S701: Write the target power mode, motor target torque, engine target torque, candidate start-up torque and sliding friction heat input ratio corresponding to the target power mode into the single mode switching execution frame; The mode switching execution frame is a single data frame containing all execution instructions such as target mode, engine target, motor target, clutch pressure rise, and timestamp, representing all mode switching instructions within a control cycle.

[0087] Step S702: Encode the single mode switching execution frame according to the communication protocol based on the synchronization timestamp to generate the engine governor target, motor inverter torque target, clutch hydraulic pressure climb command and propeller shaft side torque tracking timestamp. Among them, the engine governor target, the motor inverter torque target, and the clutch hydraulic pressure ramp-up command share the same effective timestamp, and the torque command direction is consistent with the positive direction of the propulsion shaft. The engine governor targets the target torque or equivalent speed control command sent to the engine control side to control the engine output shaft-side torque.

[0088] The motor inverter torque target is the target torque command sent to the motor inverter, which is used to control the motor output or absorb shaft-side torque.

[0089] The clutch hydraulic pressure ramp command is a command that controls the change of clutch engagement pressure over time, and is used to control the clutch engagement or disengagement process.

[0090] The propulsion shaft-side torque tracking timestamp is a time stamp used to ensure that engine, motor, and clutch commands take effect synchronously under the same execution time base, characterizing the unified effective time of the commands.

[0091] Step S703: Output the single mode switching execution frame, which is encoded according to the synchronization timestamp and the communication protocol, as a mode switching instruction so that the engine governor target, the motor inverter torque target, and the clutch hydraulic pressure ramp-up instruction take effect synchronously.

[0092] Furthermore, the mode switching execution frame must adhere to a unified communication protocol format. The execution frame includes fields such as a frame header, mode word, engine target torque, motor target torque, clutch pressure target, timestamp, and checksum. The data length is determined by the number of fields and the data type of each field, typically ranging from tens to over one hundred bytes. The communication protocol can employ commonly used marine communication methods such as CAN bus or Ethernet. For example, when using the CAN 2.0B protocol, the single frame data length is 8 bytes, and the complete execution frame data can be sent via multi-frame transmission, with the mode word occupying 1 byte, each torque value occupying 2 bytes, the timestamp occupying 2 bytes, and the checksum occupying 1 byte.

[0093] The hydraulic pressure ramp-up slope of the clutch needs to be calculated based on thermal constraints. The slope should ensure that the sliding heat input during the switching process does not exceed the preset upper limit of heat input. An excessively large slope will shorten the sliding time but increase the sliding power, while an excessively small slope will prolong the sliding time. A comprehensive selection is needed to minimize the total heat input while meeting the switching time requirements. The heat input under different slopes can be calculated by establishing a clutch thermal model, and the maximum slope that satisfies the thermal constraints can be selected to shorten the switching time. For example, the optimal slope for the pressure to rise from 0.5 MPa to 2.0 MPa is calculated to be 1 MPa per second, corresponding to a sliding heat input of 80% of the preset upper limit, which satisfies both the thermal constraints and ensures the switching speed.

[0094] The accuracy of the synchronization activation timestamp must be consistent with the data acquisition timestamp; the accuracy of the execution command activation timestamp should not exceed 10% of the control cycle, the same as the accuracy requirement of the data acquisition timestamp, to ensure the time consistency of the entire control closed loop. All execution commands share the same activation timestamp, ensuring that the engine, motor, and clutch begin executing their respective actions at the same time. Synchronization can be achieved through a bus synchronization mechanism or a hardware trigger signal. For example, using the CAN bus synchronization message mechanism, all nodes update their outputs simultaneously upon receiving the synchronization message, with time deviations controlled within 1 millisecond.

[0095] It should be noted that the mechanical capacity reflected in the mode switching load needs to be translated into synchronous execution commands that can drive the hardware before it can be implemented. If the actions of the engine, motor, and clutch are not synchronized in time, it will cause fluctuations in the propulsion torque during the switching process. For example, if the engine command takes effect 200 milliseconds earlier than the motor command, the engine torque will have already begun to change before the motor torque is established, which may lead to fluctuations in the propulsion shaft torque. This embodiment binds all commands to the same timestamp, ensuring that the engine, motor, and clutch start executing their respective actions at the same moment, achieving coordinated cooperation among the three, and guaranteeing the smoothness of the mode switching process and the continuity of propulsion torque from the execution level.

[0096] In one embodiment of the present invention, such as Figure 2 As shown, the ship hybrid power mode adaptive switching system based on operating condition identification includes: The operation data synchronization module is used to bind the collected navigation status data and shaft mechanical measurement data to the same timestamp to generate synchronized operation data; The working condition feature extraction module is used to perform hydrodynamic normalization calculation and feature distance measurement using synchronous operation data, and to extract the propeller absorbed torque, the current identified working condition and the working condition confidence level. The dynamic delay determination module is used to perform dynamic delay determination and boundary constraint analysis on various candidate power modes, together with the propeller absorbed torque and the current identified working condition, to generate propeller predicted torque and candidate establishment torque. The switching capacity calculation module is used to perform deviation time-domain integration and slip energy calculation on the propeller predicted torque, candidate establishment torque, operating condition confidence, and clutch slip state corresponding to each candidate power mode to obtain the mode switching capacity of each candidate power mode. The power mode generation module is used to compare the maximum value of each candidate power mode with the corresponding mode switching capacity, filter out the candidate power mode corresponding to the maximum value, and generate the target power mode. The target torque output module is used to substitute the target power mode and the propeller predicted torque into the preset available torque boundary for error allocation and amplitude limiting calculation, and calculate the output target torque of the motor and the target torque of the engine respectively. The switching command output module is used to encode the target power mode, motor target torque and engine target torque according to the communication protocol based on the synchronization timestamp, and generate the mode switching command output.

[0097] like Figure 3As shown, the ship's hybrid propulsion system is located in the stern and engine room. The propeller is positioned on the outer side of the stern and is connected to the gearbox, clutch, electric motor, and diesel engine via the propulsion shaft. The battery pack is located in the engine room and communicates with the main controller. The main controller establishes control or status acquisition relationships with the gearbox, clutch, electric motor, diesel engine, and battery pack. A shaft speed sensor is installed on the side of the propulsion shaft, and a clutch status sensor is installed near the clutch to collect the mechanical status of the shaft system. This structure can intuitively show the spatial relationship between the power source, transmission components, energy storage components, and control components within the hull, providing a hardware foundation for subsequent operating condition identification, mode switching, and target torque output.

[0098] like Figure 4 As shown, a propeller and propulsion shaft are located at the stern of the hull. An inflow velocity sensor and a pitch status sensor are arranged near the propeller, while a shaft speed / torque sensor is mounted on the propulsion shaft. A speed difference sensor and a pressure sensor are located at the clutch. The battery pack outputs battery status data, and the motor controller and engine controller output status data for their respective actuators. All the aforementioned navigation status data, shaft mechanical data, battery status data, and control unit data are converged to a central controller / time synchronization controller, which binds them according to the same timestamp. This structure can reflect the spatial acquisition location and convergence path of data from different sources, reducing the time deviation between operational condition identification data and shaft mechanical status.

[0099] like Figure 5 As shown, in engine propulsion mode, the diesel engine serves as the primary power source, outputting propulsion torque. The clutch is engaged, creating a continuous mechanical transmission path between the diesel engine output, the gearbox, and the propulsion shaft system. The mechanical power output from the diesel engine is transmitted to the propulsion shaft via the clutch and gearbox, which then drives the propeller. In this mode, the electric motor does not provide propulsion torque, and the battery pack and electric motor can maintain a status monitoring or standby connection. This mode effectively demonstrates the linear power transmission relationship between the diesel engine, clutch, gearbox, propulsion shaft, and propeller, making it suitable for stable cruising and other high-efficiency diesel engine navigation conditions. It also facilitates the controller in maintaining stable output based on propulsion load.

[0100] It should be noted that this invention can be deployed in the main controller of a marine hybrid power system, running as an important algorithm for mode switching management. In terms of hardware implementation, a commonly used embedded controller for ships can be used, interacting with the engine control system, motor control system, battery management system, and various sensors via CAN bus or industrial Ethernet. Regarding sensor placement, speed and torque sensors are installed on the propulsion shaft, speed and pressure sensors are installed in the gearbox or clutch location, and ship speed information is acquired from the speedometer or Doppler log, forming a data acquisition layer in conjunction with pitch angle sensors, attitude sensors, and other devices. In terms of software implementation, this invention can be integrated as an independent functional module into the ship's energy management system, working collaboratively with the underlying control algorithm, fault diagnosis module, and human-machine interface. Regarding parameter calibration, during the ship's commissioning phase, actual data from each mode switching process is collected through mooring and sea trials to calibrate key parameters such as switching duration, mode lag time, and torque build-up rate boundary, and to establish a preset operating condition sample matrix and mode switching parameter table to adapt to the specific power system characteristics of the ship.

[0101] The final output of this invention is a mode switching execution frame, which includes the target power mode, engine governor target, motor inverter torque target, clutch hydraulic pressure ramp command, and propulsion shaft torque tracking timestamp. These output commands are sent to each actuator via the ship's internal communication bus and are executed synchronously by the engine control system, motor control system, and clutch hydraulic system at a unified timestamp. For example, when the ship transitions from a stable cruising condition to an acceleration condition, the switching capacity of each candidate mode is first calculated, and the hybrid boost mode is selected as the target power mode. Then, the target torques of the engine and motor are calculated separately, and finally, a mode switching execution frame containing all execution commands is generated and output. Upon receiving the target torque command, the engine governor adjusts the fuel injection quantity; upon receiving the target torque command, the motor inverter adjusts the motor output; and upon receiving the pressure ramp command, the clutch hydraulic system adjusts the engagement pressure. All three begin operating at the same timestamp, jointly completing a smooth transition from engine propulsion mode to hybrid boost mode. After the switch is completed, the system enters a new stable operating state, continuously monitoring navigation parameters and awaiting the next mode switching determination.

[0102] The content of this embodiment has been described above, but this embodiment is not limited to the specific implementation methods described above. The specific implementation methods described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms based on the guidance of this embodiment, all of which are within the protection scope of this embodiment.

Claims

1. A method for adaptive switching of ship hybrid power modes based on operating condition recognition, characterized in that, Includes the following steps: Step S1: Bind the collected navigation status data and shaft mechanical measurement data to the same timestamp to generate synchronized operation data; Step S2: Utilize synchronized operation data to perform hydrodynamic normalization calculations and feature distance measurements to extract propeller absorbed torque, current identified operating condition, and operating condition confidence level. Step S3: Perform dynamic delay determination and boundary constraint analysis on each candidate power mode together with the propeller absorbed torque and the current identified working condition to generate the propeller predicted torque and candidate establishment torque. Step S4: Perform time-domain integral and friction energy calculation on the propeller predicted torque, candidate start-up torque, operating condition confidence level, and clutch slippage state corresponding to each candidate power mode to obtain the mode switching capacity of each candidate power mode; where: The absolute deviation between the propeller predicted torque and the candidate established torque is integrated over the same switching duration, and then normalized using the propulsion shaft rated torque and the switching duration to obtain the torque phase misalignment. The absolute value of the product of the clutch slip torque and the clutch slip angular velocity corresponding to each candidate power mode is integrated and normalized using a preset upper limit of heat input to obtain the slip heat input ratio. The operating condition confidence level, torque phase misalignment, and sliding friction heat input ratio are subjected to exponential decay calculation to obtain the mode switching capacity of each candidate power mode; Step S5: Compare the maximum values ​​of each candidate power mode with the corresponding mode switching capacity, filter out the candidate power mode corresponding to the maximum value, and generate the target power mode. Step S6: Substitute the target power mode and the propeller predicted torque into the preset available torque boundary for error allocation and amplitude limiting calculation, and calculate the output motor target torque and engine target torque respectively; Step S7: Encode the target power mode, motor target torque and engine target torque according to the communication protocol based on the synchronization timestamp, and generate a mode switching command output.

2. The working condition recognition based adaptive switching method of ship hybrid power modes according to claim 1, characterized in that, Collect navigation status data and shafting mechanical measurement data within the same control cycle; The collected data is subjected to anti-aliasing filtering, and the engine-end torque, motor-end torque, and clutch-end torque are converted to the propulsion shaft side. The propeller shaft speed, propeller shaft torque, propeller inflow velocity, pitch status, speed difference between the two ends of the clutch, clutch pressure, engine available torque boundary, motor available torque boundary, and battery available torque status are bound to the same timestamp to generate synchronized operating data.

3. The method of claim 1, wherein, The propeller inflow velocity, propeller shaft speed and propeller diameter in the synchronous operation data are hydrodynamically normalized to obtain the propeller advance coefficient. By substituting the propeller advance rate coefficient and pitch state into the torque coefficient graph, the propeller torque coefficient is obtained. Then, the propeller torque coefficient, water density of the navigation area, propeller diameter, and propeller shaft speed are used to calculate the torque absorbed by the propeller.

4. The working condition recognition based adaptive switching method of ship hybrid power modes according to claim 3, characterized in that, The propeller absorbed torque, torque filtering change rate, propeller inflow velocity, bow angular velocity amplitude, and clutch slip amplitude are normalized into a working condition identification vector. The feature distance between the working condition identification vector and each preset working condition sample matrix is ​​measured to obtain the current identified working condition and the working condition confidence level.

5. The method of claim 1, wherein, Based on the current identification conditions, read the switching duration, mode lag time, and torque of each candidate power mode to establish a rate boundary; The propeller absorbed torque and torque filtering change rate are predicted in the time domain to generate the propeller predicted torque. Starting with the propulsion shaft torque, dynamic delay determination and boundary constraint analysis are performed on each candidate power mode to generate candidate start-up torque; The candidate power modes include pure electric propulsion mode, engine propulsion mode, hybrid boost mode and engine charging mode, and also include the current operating mode.

6. The method of claim 1, wherein, The candidate mode set is determined based on the current identification operating condition, the battery available torque status, the engine available torque boundary, the motor available torque boundary, and the clutch engagement available status. For candidate power modes whose torque envelope cannot cover the predicted torque endpoint value of the propeller, the corresponding mode switching capacity is set to a constant of 0. The maximum value of the mode switching capacity corresponding to each candidate power mode in the candidate mode set is compared to obtain the candidate power mode corresponding to the maximum value; if at least two candidate power modes have the same mode switching capacity within a preset numerical precision, the target power mode is generated according to the preset mode sequence.

7. The method of claim 1, wherein, Read the motor participation coefficient and engine participation coefficient according to the target power mode, and determine the motor available torque boundary based on the battery available torque status; The difference between the predicted propeller torque endpoint and the engine efficiency torque corrected by the engine participation coefficient is substituted into the available torque boundary of the motor for amplitude limiting calculation to obtain the target torque of the motor. The difference between the predicted propeller torque endpoint and the motor target torque is substituted into the engine's available torque boundary for amplitude limiting calculation to obtain the engine target torque; where both the motor target torque and the engine target torque are propulsion shaft side torques.

8. The adaptive switching method for ship hybrid power mode based on operating condition identification according to claim 1, characterized in that, Write the target power mode, motor target torque, engine target torque, candidate start-up torque and sliding heat input ratio corresponding to the target power mode into the single mode switching execution frame; The single-mode switching execution frame is encoded according to the communication protocol based on the synchronization timestamp to generate the engine governor target, the motor inverter torque target, the clutch hydraulic pressure climb command and the propulsion shaft side torque tracking timestamp. Among them, the engine governor target, the motor inverter torque target, and the clutch hydraulic pressure ramp command share the same effective timestamp, and the torque command direction is consistent with the positive direction of the propulsion shaft. The single mode switching execution frame, which is encoded according to the synchronization timestamp, is output as the mode switching instruction so that the engine governor target, the motor inverter torque target, and the clutch hydraulic pressure ramp-up instruction take effect synchronously.

9. A ship hybrid power mode adaptive switching system based on operating condition identification, characterized in that, The method for adaptive switching of ship hybrid power modes based on operating condition identification as described in any one of claims 1 to 8 includes: The operation data synchronization module is used to bind the collected navigation status data and shaft mechanical measurement data to the same timestamp to generate synchronized operation data; The working condition feature extraction module is used to perform hydrodynamic normalization calculation and feature distance measurement using synchronous operation data, and to extract the propeller absorbed torque, the current identified working condition and the working condition confidence level. The dynamic delay determination module is used to perform dynamic delay determination and boundary constraint analysis on each candidate power mode, together with the propeller absorbed torque and the current identified working condition, to generate the propeller predicted torque and candidate establishment torque. The switching capacity calculation module is used to perform deviation time-domain integration and slip energy calculation on the propeller predicted torque, candidate establishment torque, operating condition confidence, and clutch slip state corresponding to each candidate power mode to obtain the mode switching capacity of each candidate power mode. The power mode generation module is used to compare the maximum value of each candidate power mode with the corresponding mode switching capacity, filter out the candidate power mode corresponding to the maximum value, and generate the target power mode. The target torque output module is used to substitute the target power mode and the propeller predicted torque into the preset available torque boundary for error allocation and amplitude limiting calculation, and calculate the output target torque of the motor and the target torque of the engine respectively. The switching command output module is used to encode the target power mode, motor target torque and engine target torque according to the communication protocol based on the synchronization timestamp, and generate the mode switching command output.

Images