A ship propulsion power limiting method and system based on multi-signal fusion

Abstract

The present application belongs to the technical field of ship electric propulsion, and particularly relates to a ship propulsion power limiting method and system based on multi-signal fusion, comprising the following steps: collecting the generator output, bus voltage and propeller speed of each generator, and obtaining the power of the previous period; in response to a generator tripping signal, extracting the change trend of the generator output of each operating generator, determining the confidence weight and lag correction amount of each operating generator according to the change trend, compensating the corresponding generator output by using the lag correction amount to obtain the corrected output of each operating generator; determining a safety coefficient based on the drop of the bus voltage, and updating the safety coefficient by using the mean value of the confidence weight of each operating generator to obtain a correction coefficient. The present application eliminates the response delay blind area, takes into account the safety margin and continuity of the control command, and effectively prevents the occurrence of unit interlocking tripping and ship power loss accidents.

Description

Technical Field

[0001] This invention belongs to the field of ship electric propulsion technology, specifically relating to a ship propulsion power limiting method and system based on multi-signal fusion. Background Technology

[0002] With the continuous development of ship size and electrification technology, electric propulsion systems equipped with dual programmable logic controller (PLC) architectures have become widely used in modern ships. Under typical full-speed or high-power propulsion conditions, the ship's power grid load is extremely high, with propulsion power typically accounting for over 70% of the total installed capacity. Under such high-load operation, if one or more generators suddenly trip due to overload protection or short-circuit faults, a significant instantaneous active power shortage will occur on the ship's main busbar. Therefore, the system must have the ability to provide accurate available power limits to the propulsion station within a very short time to quickly reduce the propulsion load and prevent cascading overload trips of remaining generators. This is a crucial industrial requirement for ensuring the ship's power supply and maintaining power and heading control.

[0003] To address the power shortage caused by sudden generator tripping, the most common traditional technique in the industry is a stepped power limiting strategy based on static thresholds and Boolean logic. In this approach, upon receiving a hard-wired signal indicating circuit breaker tripping, the power management system typically reads a snapshot of the static output of each operating generator at the moment of tripping and uses a black-and-white logic to determine the grid status. Subsequently, the system calculates an absolute power limiting command based on a pre-set fixed ratio or static stepped curve and sends it to the propulsion inverter via the fieldbus network, thereby forcibly reducing the propeller's input power to maintain the energy supply and demand balance of the remaining grid.

[0004] However, detection methods based on static snapshots and simple Boolean logic completely ignore the dynamic changes in generator output, making them highly susceptible to interference from sudden electromagnetic spikes and noise, leading to misjudgments. Furthermore, they lack a flexible compensation mechanism for data acquisition lag errors, resulting in low fidelity power limiting commands and frequent torque fluctuations in the inverter. More critically, from the generator tripping interruption trigger and the control system's calculations to the transmission of the command to the propulsion actuator via the communication link, there is an unavoidable delay of tens of milliseconds in calculation and communication. Within this control blind zone, the propulsion system maintains its high power output as before the fault, which can easily cause a destructive cascading trip of the already vulnerable remaining generator network due to failure to unload power in time, thus failing to fundamentally prevent a serious accident involving a complete power outage on the ship. Summary of the Invention

[0005] This invention provides a method and system for limiting ship propulsion power based on multi-signal fusion, in order to solve the technical problems of the lack of state lag compensation mechanism and the vulnerability to communication delay blind spots that lead to power loss of the entire ship.

[0006] In a first aspect, the present invention provides a method for limiting ship propulsion power based on multi-signal fusion, comprising the following steps: Collect the generator output, bus voltage, and propeller speed of each generator, and obtain the power from the previous period; In response to the generator trip signal, the change trend of the generator output of each operating generator is extracted, and the confidence weight and lag correction amount of each operating generator are determined accordingly. The lag correction amount is used to compensate the corresponding generator output to obtain the corrected output of each operating generator. The safety factor is determined based on the bus voltage drop. The safety factor is updated using the mean of the confidence weights of each operating generator to obtain the correction factor. The available power is obtained based on the correction factor and each corrected output. During the calculation waiting period after the generator trip signal is triggered, the pre-limit torque is obtained using the previous power and propeller speed and is issued first; after the available power is obtained, the accurate torque is obtained using the available power and propeller speed and is issued.

[0007] Its effects are as follows: extracting multi-scale output change trends to obtain lag correction, updating the voltage safety factor in combination with weighted average, and prioritizing the issuance of estimated pre-limit torque in the calculation blind zone, realizing flexible compensation for data lag error, making power limiting accurate while taking into account safety margin and grid dynamic stability, completely covering the communication delay risk window, and avoiding the cascading overload tripping of remaining units caused by failure to limit in time.

[0008] Furthermore, the changing trends of generator output for each operating generator are extracted, and the confidence weights and lag repair values ​​for each operating generator are determined accordingly, including: For any operating generator, the elapsed time from the start of the current scan cycle to the trigger time of the generator trip signal is obtained, and the ratio of the elapsed time to the preset scan cycle constant is used as the residual progress. Extract the difference in generator output between adjacent periods as the single-period increment; Least squares linear fitting is performed using generator output from multiple consecutive cycles including the current cycle and historical cycles, and the extracted fitting slope is used as the periodic rate of change. The confidence weights are determined based on the single-period increment and the periodic variability, and the lag adjustment is determined in conjunction with the residual progress.

[0009] Furthermore, confidence weights are determined based on single-period increments and periodic variability, and lag adjustments are determined in conjunction with residual progress, including: By comparing the absolute value of the single-period increment with the absolute value of the periodic rate of change, the minimum of the two is used as the numerator, and the maximum of the two is added to the preset zero-prevention small amount as the denominator. In response to the same sign of the single-period increment and the periodic rate of change, the direction discrimination term is set to 1, and otherwise set to 0; Calculate the quotient of the numerator and denominator, and multiply the quotient by the direction discrimination term to obtain the confidence weight; The progress difference between the preset value 1 and the residual progress is calculated. The confidence weight, periodic variability and progress difference are multiplied together to obtain the lag adjustment, so as to achieve flexible compensation for time lag error.

[0010] Its effects are as follows: by comparing the extreme values ​​of single-period increment and periodic rate of change to obtain the quotient, and by combining the direction discrimination term to determine the confidence weight and calculate the expected lag correction, it overcomes the black-and-white misjudgment of traditional Boolean logic when facing complex industrial electromagnetic noise, effectively suppresses the interference of sudden peak sampling data on power prediction, and greatly improves the fidelity and anti-disturbance capability of reconstructing the true output state of the generator.

[0011] Furthermore, the safety factor is determined based on the bus voltage drop, including: Calculate the voltage difference between the preset bus rated voltage constant and the bus voltage, and use the ratio of the voltage difference to the bus rated voltage constant as the bus voltage drop. In response to a drop of less than the first threshold, the safety factor is set to a value of 1; In response to a drop that is not less than a first threshold and less than a second threshold, the excess drop amount between the drop and the first threshold is calculated, and the difference between the preset value 1 and the excess drop amount is calculated. The difference is used as a safety factor. In response to a drop of not less than the second threshold, the safety factor is set to a preset hard lower limit value.

[0012] Furthermore, the safety factor is updated using the mean of the confidence weights of each operating generator to obtain the correction factor, including: Obtain the arithmetic mean of the confidence weights of each operating generator, and use it as the mean. The mean is used to physically characterize the stability of the overall power output of the current power grid. Calculate the stability gap value between the preset value 1 and the mean value, and multiply the stability gap value by the preset penalty intensity coefficient to obtain the penalty amount; Calculate the remaining proportional factor between the preset value 1 and the penalty amount, and multiply the safety factor by the remaining proportional factor to obtain the correction factor.

[0013] Its effect is as follows: the average value of the confidence weight of each generator is used to physically characterize the overall stability of the power grid, and the safety factor is scaled twice by combining the stability gap value and the penalty amount. The output fluctuation of local units is intelligently integrated into the global state evaluation index. Under extremely unstable conditions, a smooth additional power downpressure is automatically applied, which not only maintains the continuity of control commands, but also prevents ship propulsion stall caused by excessive rigidity limiting.

[0014] Furthermore, the usable power is obtained based on the correction coefficients and the corrected outputs, including: The corrected output of each operating generator is numerically summed to obtain the total available output of the current power grid. The available power is obtained by multiplying the correction factor by the total available output.

[0015] Furthermore, the pre-limited torque is obtained using the previous power and propeller speed, including: Multiply the previous power by the preset conservative coefficient and the propulsion transmission efficiency constant to obtain the blind zone estimated power. The propeller speed is converted into propeller angular velocity, and the ratio of the estimated power in the blind zone to the propeller angular velocity is used as the limiting torque to prevent high power output during communication delays.

[0016] Its effect is as follows: In the sudden, extremely short calculation blind zone, by combining the previous power with the conservative coefficient and the transmission efficiency constant, the pre-limit torque is quickly derived and prioritized for distribution, successfully constructing the first-stage rapid physical protection barrier. Within microseconds, the overload risk of maintaining high power output during the communication waiting window is resolved, effectively ensuring the hardware safety of the underlying electric propulsion equipment.

[0017] Furthermore, precise torque is obtained using available power and propeller speed, including: The accurate target power is obtained by multiplying the available power by the propulsion transmission efficiency constant; The propeller speed is converted into propeller angular velocity, and the ratio of the precise target power to the propeller angular velocity is used as the precise torque.

[0018] Furthermore, before collecting the generator output, bus voltage, and propeller speed of each generator, and obtaining the previous period's power, the control method also includes: During the normal scanning cycle when no generator trip signal is triggered, the active power output and bus voltage of each generator are continuously collected, and the collected values ​​are rolled into the status cache using a first-in-first-out data structure to form a sliding history window. The system synchronously collects the propeller speed reported by the frequency converter and the available power generated in the previous cycle, and stores them in the local register for continuous refreshing.

[0019] Its effects are as follows: During the normal and non-tripping regular scanning cycle, parameters are continuously collected and stored in the cache and a sliding window is formed in a first-in-first-out structure. At the same time, the local register is refreshed synchronously, realizing the normalized storage of electromechanical core computing data. In the event of a sudden hardware interruption due to a trip, the system communication time for temporary data reading is completely eliminated, and the control response speed is brought close to the hardware limit, providing high-freshness data support for the subsequent rapid and high-frequency extraction of state trends.

[0020] Secondly, the present invention provides a ship propulsion power limiting system based on multi-signal fusion, including a memory and a processor. The memory stores computer program instructions, and when the computer program instructions are executed by the processor, the above-mentioned ship propulsion power limiting method based on multi-signal fusion is implemented.

[0021] The beneficial effects are: by pre-stored electromechanical equipment data, combined with the output change trend for lag compensation, and based on the dual assessment of voltage drop and overall stability, the available power is calculated, and finally, a two-stage torque limit command is distributed. This fundamentally eliminates the control blind spot caused by response delay, effectively resists the risk of ship-wide power loss caused by extreme power shortages, and ensures the stability and survivability of the ship's propulsion system in the event of severe sea conditions and sudden tripping. Attached Figure Description

[0022] Figure 1 This is a flowchart of a ship propulsion power limiting method based on multi-signal fusion.

[0023] Figure 2 A comparison curve of the torque delivery response process.

[0024] Figure 3 The time lag error degradation curves of the method of the present invention and traditional Boolean logic control are shown. Detailed Implementation

[0025] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0026] An embodiment of the ship propulsion power limiting method based on multi-signal fusion provided by this invention: like Figure 1 As shown, the ship propulsion power limiting method based on multi-signal fusion includes the following steps: S101 collects the generator output, bus voltage, and propeller speed of each generator, and obtains the power from the previous period.

[0027] In one embodiment, to eliminate the time consumption and waiting delay of temporary data acquisition during sudden tripping interruptions, data pre-storage and cache updates are required under normal conditions. Before acquiring the generator output, bus voltage, and propeller speed of each generator and obtaining the power from the previous period, the control method further includes: continuously acquiring the active power output and bus voltage of each generator during the normal scanning cycle when no generator tripping signal is triggered, and using a first-in-first-out data structure to continuously store the acquired values ​​into the status cache to form a sliding history window; synchronously acquiring the propeller speed reported by the frequency converter and the available power generated in the previous cycle, and storing them in the local register for continuous refreshing.

[0028] In this context, the active power output mentioned above refers to the generator output of each generator, and the available power generated in the previous cycle is the power from the previous period used for the current cycle's control calculations. The power management station continuously collects the generator output and bus voltage of each generator on the ship's main bus during normal scanning cycles, and continuously stores the collected results in the status cache. For any given generator, at the end of the current scanning cycle, the current sampled value of its generator output is stored in the status cache, while sampled values ​​from multiple historical cycles are also retained in the status cache. An in-first-out (FIFO) data structure is used to iteratively update the sliding historical window. Simultaneously, the sampled value of the bus voltage is also stored in the status cache.

[0029] Preferably, the scan cycle constant is 10 milliseconds, and the length of the sliding history window is 4 cycles, meaning the sliding history window includes the sampled value of the current cycle and the sampled values ​​of the past three consecutive historical cycles, for a total of four data points. This length can completely cover the power reduction process from 30 to 50 milliseconds before the generator trips. During the on-site commissioning phase, the above optimal value is obtained by statistically analyzing the trip waveform data to balance data freshness and anti-interference capability. While the power management station performs the above operations, the propulsion station continuously collects the propeller speed reported by the propulsion inverter within the normal scan cycle and stores it in the local register, and synchronously stores the previous period's power in the local register for continuous refreshing. Based on the above pre-storage mechanism, at any time when a trip interruption is triggered, the basic data collection and acquisition can be completed with zero delay by directly extracting the generator output and bus voltage of each generator from the status cache, and extracting the propeller speed and previous period's power from the local register.

[0030] Within a very short time after a sudden trip causes a hardware interruption, all the latest status data and recent historical data required for calculation are pre-stored in the status cache and local registers. This fundamentally eliminates the time consumption of temporary data acquisition during the interrupt response process, avoids the obstruction caused by communication link delays to the underlying actuators, and provides complete and reliable data support for the subsequent rapid calculation of generator dynamic change trends and the implementation of power limits.

[0031] S102, in response to the generator trip signal, extracts the changing trend of the generator output of each operating generator, determines the confidence weight and lag correction amount of each operating generator, and uses the lag correction amount to compensate the corresponding generator output to obtain the corrected output of each operating generator.

[0032] In one embodiment, when a sudden generator trip signal is detected, the control process immediately suspends the regular scan task and triggers a highest-priority hardware interrupt. To quantify output trends and provide a time conversion benchmark across dual time scales, the output change trends of each operating generator are extracted, and the confidence weights and hysteresis corrections for each operating generator are determined accordingly. This includes: for any operating generator, obtaining the elapsed time from the start of the current scan cycle to the trigger time of the generator trip signal, and using the ratio of the elapsed time to a preset scan cycle constant as the residual progress; extracting the difference in generator output between adjacent cycles as a single-period increment; performing least-squares linear fitting using generator output from multiple consecutive cycles including the current and historical cycles, and using the extracted fitting slope as the periodic rate of change; determining the confidence weights based on the single-period increment and the periodic rate of change, and combining this with the residual progress to determine the hysteresis correction.

[0033] Among them, the residual progress is the quotient obtained by dividing the elapsed time by the scan cycle constant, and its value ranges from 0 to 1. It can characterize the proportion of time occupied by the interruption trigger moment in the current scan cycle; the single-period increment is the arithmetic difference between the generator output of the current cycle and the generator output of the previous adjacent cycle, which is used to reflect the instantaneous transient change in power; the cycle rate is the slope of the best fitted line extracted from the generator output of four consecutive cycles in the state buffer using the standard least squares linear fitting algorithm, which represents the smooth output change in each scan cycle.

[0034] In another embodiment, a confidence weight is determined based on the single-period increment and the periodic rate of change, and a lag correction is determined in conjunction with the residual progress. This includes: comparing the absolute value of the single-period increment with the absolute value of the periodic rate of change, taking the minimum of the two as the numerator, and adding the maximum of the two to a preset zero-prevention value as the denominator; in response to the single-period increment and the periodic rate of change having the same sign, setting the direction discrimination term to 1, and otherwise setting it to 0; calculating the quotient of the numerator and the denominator, and multiplying the quotient by the direction discrimination term to obtain the confidence weight; calculating the progress difference between the preset value 1 and the residual progress, and multiplying the confidence weight, the periodic rate of change, and the progress difference to obtain the lag correction, so as to achieve flexible compensation for time lag error.

[0035] The confidence weights satisfy the following relationship:

[0036] In the formula, As confidence weights, and These are functions for finding the minimum and maximum values, respectively. This represents the absolute value of the increment in a single period. The absolute value of the periodic rate of change. To prevent small quantities, This is the direction discrimination term. Preferably, the value of the zero-prevention quantity is 0.1 kW, which is chosen to prevent the denominator from returning to zero when the generator output is extremely stable, thus causing the calculation to crash.

[0037] Understandably, the closer the instantaneous change amplitude of a single period is to the long-term change amplitude of multiple periods, the closer the ratio of the two is to 1. This indicates that the current output change trend is extremely stable and has not been affected by sudden hardware noise, and is automatically given a very high degree of correction confidence. Conversely, if the two values ​​are significantly different or opposite in direction, resulting in the direction discrimination term being zero, the confidence weight will automatically and smoothly decay to a safe low level.

[0038] Based on the determined confidence weights, further flexible compensation is performed, and the lag adjustment satisfies the following relationship:

[0039] In the formula, To revise the quantity in a delayed manner, As confidence weights, For periodic variation, For the remaining progress, This is the difference between the preset value of 1 and the remaining progress, representing the progress difference.

[0040] Understandably, the product of the progress difference and the cycle variability represents the expected output change over the remaining time of the current cycle. This expected change is then scaled for confidence using a confidence weight to output a compensation value with extremely high safety fidelity. Finally, by directly adding the lag correction to the generator output of the current cycle, the corrected output of each operating generator can be obtained.

[0041] In the early stages of generator tripping, cross-verification of time conversion and output trends under dual time scales was completed, and the measurement error introduced by the scanning cycle lag was flexibly compensated based on physical fit. This step not only ensured the absolute correctness of the correction direction but also effectively suppressed the interference of sudden electromagnetic spikes on power prediction, improving the anti-disturbance capability of the reconstructed state data.

[0042] S103: Determine the safety factor based on the bus voltage drop, update the safety factor using the average of the confidence weights of each operating generator to obtain the correction factor, and obtain the available power based on the correction factor and each corrected output.

[0043] In one embodiment, to eliminate frequent jumps in available power limit commands caused by minor voltage fluctuations near a critical threshold and to quantify the upper limit of the safe power actually available to the entire ship, a safety factor is determined based on the bus voltage drop, including: calculating the voltage difference between a preset bus rated voltage constant and the bus voltage, and using the ratio of the voltage difference to the bus rated voltage constant as the bus voltage drop; in response to the drop being less than a first threshold, setting the safety factor to a value of 1; in response to the drop being not less than the first threshold and less than a second threshold, calculating the drop excess between the drop and the first threshold, calculating the difference between the preset value 1 and the drop excess, and using the difference as the safety factor; and in response to the drop being not less than the second threshold, setting the safety factor to a preset hard lower limit value.

[0044] It should be noted that, under islanded network conditions, the voltage drop on the bus after a generator trip is a reliable external physical indicator of the ship's active power deficit. The aforementioned voltage drop is calculated by subtracting the currently measured bus voltage from the rated bus voltage constant, and then dividing by the rated bus voltage constant. To achieve a smooth response to voltage drops, the safety factor satisfies the following relationship:

[0045] In the formula, For safety reasons, For the decline, The first threshold, The second threshold, The lower limit is a hard threshold value. Preferably, the first threshold is 0.05, the second threshold is 0.15, and the hard lower limit is 0.90. The first threshold represents the initial adjustment threshold for voltage drop, the second threshold represents the full adjustment threshold for voltage drop, and the hard lower limit represents the safety limit value to prevent propulsion stall. When the voltage drop is between the first and second thresholds, the safety factor decreases linearly and smoothly. No power penalty is applied when the voltage drop is less than the initial adjustment threshold; as the voltage drop intensifies, the safety factor transitions smoothly through linear interpolation; and the hard lower limit is maintained over a larger voltage drop range to strictly prevent excessive restriction that could lead to propulsion stall.

[0046] In another embodiment, to integrate and reuse overall grid stability information into power limitation calculations and achieve dual perception of grid status, the safety factor is updated using the mean of the confidence weights of each operating generator to obtain a correction factor. This includes: obtaining the arithmetic mean of the confidence weights of each operating generator and using it as the mean, which is used to physically characterize the stability of the current overall grid output; calculating the stability gap value between a preset value 1 and the mean, and multiplying the stability gap value by a preset penalty intensity coefficient to obtain a penalty amount; calculating the remaining proportional factor between the preset value 1 and the penalty amount, and multiplying the safety factor by the remaining proportional factor to obtain the correction factor.

[0047] In step S102, the arithmetic mean of the confidence weights of all operating generators is the mean. If the output of each generator fluctuates drastically and in different directions, and the mean drops significantly, it indicates that the power grid is in a highly unstable state, requiring additional penalties to be imposed on the safety factor. To quantify the above stability indicators and perform a second scaling on the basic safety factor obtained based on the bus voltage drop, the correction coefficient satisfies the following relationship:

[0048] In the formula, For correction factor, For safety reasons, The penalty intensity coefficient, The mean, This is the stability gap value, which characterizes insufficient stability. This refers to the penalty amount calculated based on the degree of instability. This is the remaining scaling factor used for secondary scaling. Preferably, the penalty intensity coefficient is 0.05. This value is based on ensuring that even under the most extreme grid instability conditions, the maximum additional decrease of the correction coefficient relative to the safety factor does not exceed 0.05, with the absolute lower limit constraint being the absence of propulsion stall.

[0049] For example, if the bus voltage drop is 0.10, the safety factor calculated according to the above logic is 0.95, and the output of each operating generator fluctuates greatly, resulting in an average of 0.6, then the correction factor is the product of 0.95 and 0.98, i.e. 0.931, which makes the power command further smoothed down on the basis of voltage drop correction, effectively dealing with the additional risks brought about by grid instability.

[0050] In another embodiment, in order to output the final safe execution benchmark based on the updated core parameters, the available power is obtained based on the correction coefficient and each correction output, including: numerically summing the correction output of each operating generator to obtain the total available output of the current power grid; and multiplying the correction coefficient by the total available output to obtain the available power.

[0051] Understandably, by algebraically summing the corrected output of all generators in operation after time lag compensation, we can obtain the total available output representing the current basic supply capacity of the power grid. Subsequently, by directly multiplying the total available output by a correction factor that has undergone dual assessment of voltage drop and grid stability, we can obtain the final available power that balances conservative safety margins and grid dynamic stability.

[0052] The calculation of available power not only has a conservative safety margin to withstand extreme power shortages and prevent the entire ship from losing power, but also ensures excellent smoothness and continuity of control commands subsequently issued to mechanical equipment by reusing the overall stability information of the power grid in the secondary correction of the safety factor.

[0053] S104: During the calculation waiting period after the generator trip signal is triggered, the pre-limit torque is obtained using the previous power and propeller speed and is issued first; after the available power is obtained, the accurate torque is obtained using the available power and propeller speed and is issued.

[0054] In one embodiment, in order to completely cover the risk window caused by communication delay, during the extremely short calculation waiting period triggered by the power station interrupt service routine, i.e., the communication dead zone of about 10 milliseconds, a pre-limited torque is obtained using the previous power and propeller speed. This includes: multiplying the previous power by a preset conservative coefficient and the propulsion transmission efficiency constant to obtain the estimated power for the dead zone; converting the propeller speed into the propeller angular velocity, and using the ratio of the estimated power for the dead zone to the propeller angular velocity as the pre-limited torque to prevent high power output during the communication delay period.

[0055] It should be noted that this stage is the blind zone pre-limitation stage, and the propulsion station immediately performs conservative calculations based on the various equipment status parameters pre-stored in step S101; the pre-limit torque satisfies the following relationship:

[0056] In the formula, To limit torque, The power from the previous period is obtained from the local register. This is a conservative coefficient. To advance the transmission efficiency constant, This is the estimated power of the blind zone after unit conversion to watts; The propeller speed is obtained from the local register. This is a fixed conversion factor from rotational speed to angular velocity. This refers to the propeller angular velocity. Preferably, the conservatism coefficient is 0.80, which physically means that the remaining usable power after the tripping of the largest single generator is approximately 0.80 times the original value. This parameter is adjusted during the commissioning phase based on the capacity ratio of the largest single generator. The typical value for the propulsion transmission efficiency constant is 0.95. The propulsion station directly writes the pre-limit torque into the torque limit register of the frequency converter and immediately sends it out to ensure that the ship's propulsion power will not remain at the high level before the fault in the communication dead zone.

[0057] After the power control station completes the full correction calculation in step S103 and completes the flag bit through digital channel flip communication, the control process enters the second stage of refinement, which uses available power and propeller speed to obtain precise torque, including: multiplying available power by the propulsion transmission efficiency constant to obtain precise target power; converting propeller speed into propeller angular velocity, and using the ratio of precise target power to propeller angular velocity as precise torque.

[0058] In this process, the propulsion station reads the latest available power generated in step S103 and immediately calculates the final execution command; the precise torque satisfies the following relationship:

[0059] In the formula, For precise torque, The available power output in step S103 To advance the transmission efficiency constant, This is the precise target power after unit conversion; The propeller speed, This refers to the propeller angular velocity. Understandably, since the rate of change of propeller speed on large ships is typically extremely small, its change is negligible during the 10-millisecond calculation wait period. Therefore, directly reusing the propeller speed pre-stored in step S101 will not introduce additional errors. Subsequently, the precise torque is updated to the inverter's torque limit register, completing the final control loop.

[0060] By employing a phased execution mechanism consisting of blind zone pre-limitation and precise correction, the risk of short-term overload within the communication waiting window is mitigated within microseconds. Furthermore, high-precision flexible power limiting is achieved after data calculation is complete, ensuring the survivability of the electric propulsion architecture under extreme failures and the long-term stability of ship maneuvering.

[0061] like Figure 2 As shown, during the calculation waiting period after the generator trip signal is triggered, traditional techniques still maintain extremely high torque, which can easily lead to overload. However, the method of this invention successfully eliminates the response delay blind zone by prioritizing the issuance of the estimated pre-limit torque, and smoothly transitions to the precise torque after a very short time, taking into account both the safety margin and continuity of the control command.

[0062] like Figure 3 As shown, the time lag error degradation curves of the method of this invention and the traditional Boolean logic control are displayed under different degrees of sudden interference. As shown in the severe operating condition range in the figure, when faced with the aggravation of complex industrial electromagnetic noise, the traditional method suffers from a sharp increase in lag error due to the lack of a flexible compensation mechanism. However, the present invention effectively suppresses the interference of sudden electromagnetic spikes on power prediction by extracting the periodic rate of change and comparing it with the single-period increment for amplitude limiting, and still maintains excellent robustness under extreme operating conditions.

[0063] An embodiment of the ship propulsion power limiting system based on multi-signal fusion provided by the present invention: The ship propulsion power limiting system based on multi-signal fusion includes a processor and a memory. The memory stores computer program instructions, which, when executed by the processor, implement the aforementioned ship propulsion power limiting method based on multi-signal fusion.

[0064] The ship propulsion power limiting system based on multi-signal fusion also includes other components well known to those skilled in the art, such as communication interfaces. Their settings and functions are known in the art and will not be described in detail here.

[0065] In this invention, the aforementioned memory can be any tangible medium containing or storing a program that can be used or combined with an instruction execution system, apparatus, or device. For example, a computer-readable storage medium can be any suitable magnetic or magneto-optical storage medium, such as Resistive Random Access Memory (RRAM), Dynamic Random Access Memory (DRAM), Static Random Access Memory (SRAM), Enhanced Dynamic Random Access Memory (EDRAM), High-Bandwidth Memory (HBM), Hybrid Memory Cube (HMC), etc., or any other medium that can be used to store desired information and can be accessed by an application, module, or both. Any such computer storage medium can be part of a device or accessible to or connected to a device. Any application or module described in this invention can be implemented using computer-readable / executable instructions stored or otherwise maintained by such a computer-readable medium.

[0066] The above are all preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Therefore, all equivalent changes made in accordance with the structure, shape and principle of the present invention should be covered within the scope of protection of the present invention.

Claims

1. A method for limiting ship propulsion power based on multi-signal fusion, characterized in that, Includes the following steps: Collect the generator output, bus voltage, and propeller speed of each generator, and obtain the power from the previous period; In response to the generator trip signal, the change trend of the generator output of each operating generator is extracted, and the confidence weight and lag correction amount of each operating generator are determined accordingly. The lag correction amount is used to compensate the corresponding generator output to obtain the corrected output of each operating generator. The safety factor is determined based on the bus voltage drop. The safety factor is updated using the mean of the confidence weights of each operating generator to obtain the correction factor. The available power is obtained based on the correction factor and each corrected output. During the calculation waiting period after the generator trip signal is triggered, the pre-limit torque is obtained using the previous power and propeller speed and is issued first; after the available power is obtained, the accurate torque is obtained using the available power and propeller speed and is issued.

2. The ship propulsion power limiting method based on multi-signal fusion according to claim 1, characterized in that, Extract the changing trends of generator output for each operating generator, and determine the confidence weight and lag repair amount for each operating generator accordingly, including: For any operating generator, the elapsed time from the start of the current scan cycle to the trigger time of the generator trip signal is obtained, and the ratio of the elapsed time to the preset scan cycle constant is used as the residual progress. Extract the difference in generator output between adjacent periods as the single-period increment; Least squares linear fitting is performed using generator output from multiple consecutive cycles including the current cycle and historical cycles, and the extracted fitting slope is used as the periodic rate of change. The confidence weights are determined based on the single-period increment and the periodic variability, and the lag adjustment is determined in conjunction with the residual progress.

3. The ship propulsion power limiting method based on multi-signal fusion according to claim 2, characterized in that, Confidence weights are determined based on single-period increments and periodic variability, and lag adjustments are determined in conjunction with residual progress, including: By comparing the absolute value of the single-period increment with the absolute value of the periodic rate of change, the minimum of the two is used as the numerator, and the maximum of the two is added to the preset zero-prevention small amount as the denominator. In response to the same sign of the single-period increment and the periodic rate of change, the direction discrimination term is set to 1, and otherwise set to 0; Calculate the quotient of the numerator and denominator, and multiply the quotient by the direction discrimination term to obtain the confidence weight; The progress difference between the preset value 1 and the residual progress is calculated. The confidence weight, periodic variability and progress difference are multiplied together to obtain the lag adjustment, so as to achieve flexible compensation for time lag error.

4. The ship propulsion power limiting method based on multi-signal fusion according to claim 1, characterized in that, The safety factor is determined based on the bus voltage drop, including: Calculate the voltage difference between the preset bus rated voltage constant and the bus voltage, and use the ratio of the voltage difference to the bus rated voltage constant as the bus voltage drop. In response to a drop of less than the first threshold, the safety factor is set to a value of 1; In response to a drop that is not less than a first threshold and less than a second threshold, the excess drop amount between the drop and the first threshold is calculated, and the difference between the preset value 1 and the excess drop amount is calculated. The difference is used as a safety factor. In response to a drop of not less than the second threshold, the safety factor is set to a preset hard lower limit value.

5. The ship propulsion power limiting method based on multi-signal fusion according to claim 1, characterized in that, The safety factor is updated using the mean of the confidence weights of each operating generator to obtain the correction factor, including: Obtain the arithmetic mean of the confidence weights of each operating generator, and use it as the mean. The mean is used to physically characterize the stability of the overall power output of the current power grid. Calculate the stability gap value between the preset value 1 and the mean value, and multiply the stability gap value by the preset penalty intensity coefficient to obtain the penalty amount; Calculate the remaining proportional factor between the preset value 1 and the penalty amount, and multiply the safety factor by the remaining proportional factor to obtain the correction factor.

6. The ship propulsion power limiting method based on multi-signal fusion according to claim 1, characterized in that, The usable power is obtained based on the correction coefficients and the corrected output power, including: The corrected output of each operating generator is numerically summed to obtain the total available output of the current power grid. The available power is obtained by multiplying the correction factor by the total available output.

7. The ship propulsion power limiting method based on multi-signal fusion according to claim 1, characterized in that, The pre-limit torque is obtained using the previous power and propeller speed, including: Multiply the previous power by the preset conservative coefficient and the propulsion transmission efficiency constant to obtain the blind zone estimated power. The propeller speed is converted into propeller angular velocity, and the ratio of the estimated power in the blind zone to the propeller angular velocity is used as the limiting torque to prevent high power output during communication delays.

8. The ship propulsion power limiting method based on multi-signal fusion according to claim 1, characterized in that, Precise torque is obtained using available power and propeller speed, including: The accurate target power is obtained by multiplying the available power by the propulsion transmission efficiency constant; The propeller speed is converted into propeller angular velocity, and the ratio of the precise target power to the propeller angular velocity is used as the precise torque.

9. The ship propulsion power limiting method based on multi-signal fusion according to claim 1, characterized in that, Before collecting the generator output, bus voltage, and propeller speed of each generator, and obtaining the previous period's power, the control method also includes: During the normal scanning cycle when no generator trip signal is triggered, the active power output and bus voltage of each generator are continuously collected, and the collected values ​​are rolled into the status cache using a first-in-first-out data structure to form a sliding history window. The system synchronously collects the propeller speed reported by the frequency converter and the available power generated in the previous cycle, and stores them in the local register for continuous refreshing.

10. A ship propulsion power limiting system based on multi-signal fusion, characterized in that, It includes a memory and a processor, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the ship propulsion power limiting method based on multi-signal fusion as described in any one of claims 1-9 is implemented.

Images