A ship maintenance system

Abstract

The present application relates to the technical field of ship maintenance, and discloses a ship maintenance system, which comprises an analysis unit, a calculation unit, a processing unit, an adaptation unit, a maintenance unit and an execution unit, is used for determining maintenance instructions according to comprehensive maintenance values, and is cooperated through the units, wherein the analysis unit captures torque signals to generate gear surface alteration potential values, the calculation unit combines temperature and vibration to obtain alteration conductivity, the processing unit analyzes oil images to obtain lubricating energy attenuation coefficients, the adaptation unit matches to generate maintenance imbalance indexes, the maintenance unit combines the use time to calculate comprehensive maintenance values, and the execution unit issues instructions accordingly, so that hidden damage can be accurately identified, the state of gears and oil can be matched, wear accumulation can be avoided, the effect of maintenance and repair can be ensured, and the maintenance requirements of the long-term stable maintenance of the ship can be met.

Description

Technical Field

[0001] This invention relates to the field of ship repair and maintenance technology, and specifically to a ship repair and maintenance system. Background Technology

[0002] As non-self-propelled vessels that are moored in fixed waters for extended periods, pontoons are mainly used for cargo loading and unloading, personnel embarkation and disembarkation, floating work platforms, or living assistance. Their maintenance and upkeep needs differ significantly from those of self-propelled vessels, such as cargo ships and cruise ships. Pontoons place greater emphasis on the long-term stability maintenance of static equipment, hull structure, mooring equipment, and ancillary facilities.

[0003] Currently, when maintaining winches in barge mooring equipment, the gear condition is usually checked during oil changes to achieve the purpose of maintenance. However, the above maintenance method still has the following defects: the timing of oil changes mainly depends on the appearance and temperature of the gear oil to determine whether it needs to be changed. Although gear wear is checked at the same time, it focuses more on the visible wear that has occurred (such as obvious scratches and tooth surface peeling), while the early hidden damage (such as slight tooth surface scuffing and abrasion) caused by the frequent switching between zero speed and peak torque of barge winches is often ignored. In this case, the protective effect of the newly changed gear oil under continuous dynamic impact does not match the actual damage progress of the gears, which eventually leads to the continuous accumulation of hidden wear, resulting in poor actual maintenance effect. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a ship repair and maintenance system that solves the aforementioned problems.

[0005] The above-mentioned technical objective of the present invention is achieved through the following technical solution:

[0006] A ship repair and maintenance system, comprising:

[0007] The analysis unit is used to acquire the torque signal, speed signal, tooth surface temperature and vibration frequency of the target object in real time during the target process. It analyzes the pre-processed torque signal and speed signal to obtain the tooth surface erosion potential value. The target object is the gear disk of the pontoon winch, and the target process is: the switching process of the winch gear disk from zero speed to peak torque.

[0008] The calculation unit is used to calculate the tooth surface alteration potential value based on the tooth surface temperature and vibration frequency, and obtain the alteration conductivity.

[0009] The processing unit is used to acquire image data of gear oil in the target object in real time, analyze the preprocessed image data, and generate a lubrication attenuation coefficient.

[0010] The adapter unit is used to match the alteration conductivity and the lubrication attenuation coefficient to obtain the maintenance imbalance index;

[0011] The maintenance unit is used to obtain the usage time of gears and gear oil in the target object in real time, and calculate the maintenance imbalance index based on the usage time of gears and gear oil to obtain the comprehensive maintenance value.

[0012] The execution unit is used to determine maintenance instructions based on the comprehensive maintenance value.

[0013] Furthermore, the preprocessed torque and speed signals are analyzed to obtain the tooth surface etching potential value, including:

[0014] The timing characteristics of torque and speed signal transformations during the target process are obtained. Based on the timing characteristics, the target process is divided into three stages: start-up, ramp-up, and peak maintenance, and the duration of each stage is calculated.

[0015] Based on the phase duration ratio, the preprocessed torque and speed signals are enhanced in stages to generate phased torque and speed feature quantities.

[0016] Based on the phased torque and speed characteristics, the correlation between torque and speed in the target process is analyzed, and a torque-speed linkage coefficient is generated.

[0017] Obtain the structural properties of the target object, combine the torque-speed linkage coefficient with the structural properties to calculate the tooth surface stress fluctuation value;

[0018] The rates of change of torque and speed are calculated from the preprocessed torque and speed signals.

[0019] By fusing the stress fluctuation value and the rate of change of the tooth surface, the tooth surface etching potential value is obtained.

[0020] Furthermore, by combining the torque-speed linkage coefficient and structural properties, the tooth surface stress fluctuation value is obtained, including:

[0021] The structural properties are analyzed, and weights are assigned according to the stress sensitivity of the target object in the target process to generate dynamic characteristic values ​​of the structure.

[0022] By analyzing the variation patterns of structural dynamic characteristic values ​​and torque-speed linkage coefficient at each stage, the structure-linkage dynamic coupling coefficient is obtained.

[0023] Based on structural properties, the deformation of the gear at different meshing positions during contact is calculated to obtain the gear contact deformation.

[0024] The structure-linkage dynamic coupling coefficient is calibrated based on the gear contact deformation to generate the tooth surface stress fluctuation value.

[0025] Furthermore, the tooth surface alteration potential is calculated based on the tooth surface temperature and vibration frequency to obtain the alteration conductivity, including:

[0026] Based on the tooth surface temperature, the frequency and duration of temperature rise points during the target process are analyzed to generate thermal shock characteristic values.

[0027] Based on the vibration frequency, the jump amplitude and duration period of abnormal vibration during the target process are extracted to obtain the vibration frequency anomaly characteristic value;

[0028] The microstructure parameters of the gear tooth surface are obtained, and the thermal shock characteristic value and the vibration frequency anomaly characteristic value are calculated based on the microstructure parameters to obtain the thermal shock index and the vibration frequency anomaly index.

[0029] The synchronicity of thermal shock index and frequency anomaly index on the time axis is analyzed to generate thermal shock synergy factor.

[0030] Furthermore, the tooth surface alteration potential is calculated based on the tooth surface temperature and vibration frequency to obtain the alteration conductivity, which also includes:

[0031] The mapping relationship between thermal shock synergy factor and tooth surface etching potential value was analyzed to obtain the synergy-etching correlation degree;

[0032] The influence of different tooth surface temperatures and vibration frequencies on the gear is calculated, and the co-alteration correlation degree is corrected based on the influence values ​​to obtain the alteration conductivity.

[0033] Furthermore, the preprocessed image data is analyzed to generate a lubrication attenuation coefficient, including:

[0034] Image data is analyzed to capture the discrete trajectory and aggregation frequency of the microstructure inside the oil, generating an indicator of oil phase dispersion.

[0035] The change in the light and shadow gradient of the oil film during gear meshing in the image data is calculated to obtain the boundary stability value of the oil film;

[0036] The correlation between oil phase dispersion indicators and oil film boundary stability values ​​during gear operation cycles is analyzed to generate performance decay coupling values.

[0037] Analyze the magnitude and speed of changes in the shape of oil in the image data as torque and speed change, and generate the working condition adaptation variation rate;

[0038] The efficiency attenuation coupling value is fused with the operating condition adaptation variation rate to obtain the lubrication attenuation coefficient.

[0039] Furthermore, by matching the alteration conductivity and the lubrication attenuation coefficient, a maintenance imbalance index is obtained, including:

[0040] The alteration conductivity and the lubrication attenuation coefficient are respectively extended in terms of feature dimension to generate alteration feature vector and lubrication feature vector;

[0041] The image data is analyzed to obtain the gear oil viscosity. The gear oil viscosity is then combined with the microtexture structure parameters to obtain the viscosity-roughness matching value.

[0042] Calculate the spatial matching degree between the alteration eigenvector and the lubrication eigenvector to obtain the alteration-lubrication matching deviation value;

[0043] The lubrication attenuation rate is calculated based on the alteration conductivity as a function of the gear tooth surface alteration, and a lubrication attenuation rate value is generated.

[0044] The viscosity-roughness fit value and the lubrication decay rate value are used as weight parameters to adjust the alteration-lubrication energy matching deviation value, thereby generating a dynamic fit imbalance.

[0045] Based on the dynamic adaptation imbalance, the peak ranges of torque and speed signals are mapped to generate a maintenance imbalance index.

[0046] Furthermore, based on the dynamic adaptation imbalance, the peak ranges of the torque and speed signals are mapped to generate a maintenance imbalance index, including:

[0047] Analyze the transient abrupt changes within the peak range of torque and speed signals to generate transient peak change characteristic values;

[0048] The interaction between transient peak change eigenvalues ​​and dynamic adaptation imbalance at each stage of the target process is analyzed to obtain the spatiotemporal coupling imbalance coefficient.

[0049] Based on image data, calculate the critical value for oil film rupture during gear meshing;

[0050] The spatiotemporal coupling imbalance coefficient is dynamically calibrated based on the oil film rupture critical value to generate the calibrated spatiotemporal imbalance value.

[0051] The spatiotemporal imbalance values ​​are converted to generate a maintenance imbalance index.

[0052] Furthermore, the maintenance imbalance index is calculated based on the service life of the gears and gear oil to obtain a comprehensive maintenance value, including:

[0053] The performance degradation of gears and gear oils over different service durations was analyzed, and the nonlinear degradation coefficients of gear aging and oil aging were generated.

[0054] The dynamic change rates of the gear aging nonlinear decay coefficient and the oil aging nonlinear decay coefficient are calculated to obtain the duration-decay synergy coefficient.

[0055] The maintenance imbalance index is adjusted in stages based on the duration-attenuation synergy coefficient to generate a dynamic maintenance estimate.

[0056] The stress fluctuation value of tooth surface, thermal shock characteristic value, vibration frequency anomaly index and oil film rupture critical value are fused to generate a multidimensional attenuation factor;

[0057] The comprehensive maintenance value is obtained by correcting the deviation of the dynamic maintenance estimate based on the multidimensional decay factor.

[0058] Furthermore, based on the comprehensive maintenance value, maintenance instructions are determined, including:

[0059] Based on the comprehensive maintenance value, the maintenance range and maintenance instructions are determined.

[0060] In summary, the present invention has the following main beneficial effects:

[0061] By dividing the zero-speed to peak torque switching process into three stages—start-up, ramp-up, and peak maintenance—and combining a 10ms sliding window to capture the temporal characteristics of torque and speed, a tooth surface wear potential value is generated through staged feature enhancement and linkage analysis. At the same time, structural attributes such as the number of gear teeth and module are incorporated to calculate the tooth surface stress fluctuation value, accurately mapping the accumulation process of hidden damage such as slight scuffing and scratches. This breaks through the limitation of traditional maintenance that can only detect visible wear, and can capture subtle signs of damage under frequent operating conditions of barge winches, preventing hidden wear from becoming more and more serious due to long-term neglect.

[0062] By jointly assessing the condition of gears and gear oil, and using the calculation unit to generate an alteration conductivity based on tooth surface temperature and vibration frequency to reflect the progress of gear damage, the processing unit generates a lubrication attenuation coefficient through oil image analysis to understand the protective capability of the gear oil. The adaptation unit matches the two to obtain a maintenance imbalance index, and the maintenance unit calculates a comprehensive maintenance value based on usage time. This achieves a dynamic correlation between damage state and lubrication performance. Compared with the traditional method of judging oil change time based on appearance, this method can understand the matching degree between the protective effect of gear oil and actual damage, avoiding the problem of mismatch between new oil change and the progress of hidden damage, making maintenance and repair more in line with actual needs.

[0063] The execution unit divides the equipment into four intervals based on the comprehensive maintenance value, corresponding to instructions for changing oil, changing gears, changing both at the same time, or not changing at all. This replaces the traditional experience-based decision-making method, preventing the waste of resources caused by over-maintenance and avoiding equipment failure caused by insufficient maintenance. It is especially suitable for the long-term stability maintenance requirements of pontoons for static equipment, effectively extending the service life of winch gears and gear oil, and reducing maintenance costs caused by the accumulation of hidden wear. Attached Figure Description

[0064] Figure 1 This is a schematic diagram of a ship repair and maintenance system according to the present invention. Detailed Implementation

[0065] 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 embodiments of the present invention, and not all embodiments. 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.

[0066] refer to Figure 1 A ship repair and maintenance system, comprising:

[0067] The analysis unit is used to acquire the torque signal, speed signal, tooth surface temperature and vibration frequency of the target object in real time during the target process. It analyzes the pre-processed torque signal and speed signal to obtain the tooth surface erosion potential value. The target object is the gear disk of the pontoon winch, and the target process is: the switching process of the winch gear disk from zero speed to peak torque.

[0068] The calculation unit is used to calculate the tooth surface alteration potential value based on the tooth surface temperature and vibration frequency, and obtain the alteration conductivity.

[0069] The processing unit is used to acquire image data of gear oil in the target object in real time, analyze the preprocessed image data, and generate a lubrication attenuation coefficient.

[0070] The adapter unit is used to match the alteration conductivity and the lubrication attenuation coefficient to obtain the maintenance imbalance index;

[0071] The maintenance unit is used to obtain the usage time of gears and gear oil in the target object in real time, and calculate the maintenance imbalance index based on the usage time of gears and gear oil to obtain the comprehensive maintenance value.

[0072] The execution unit is used to determine maintenance instructions based on the comprehensive maintenance value.

[0073] By analyzing signals such as torque and speed during the zero-speed to peak torque switching process in real time, the analysis unit accurately identifies hidden damage such as minor tooth surface scuffing. The calculation unit combines tooth surface temperature and vibration frequency to generate the alteration conductivity, and the processing unit obtains the lubrication energy attenuation coefficient based on gear oil image data. After matching by the adaptation unit, a maintenance imbalance index is formed, which enables dynamic correlation analysis between gear damage and lubricating oil protection capability. The maintenance unit calculates a comprehensive maintenance value based on usage time, which makes the maintenance instructions issued by the execution unit more in line with actual needs, avoiding the problem of mismatch between new gear oil replacement and the progress of hidden gear damage, significantly improving the maintenance effect of the pontoon winch and extending the service life of the equipment.

[0074] In one embodiment, the preprocessed torque and speed signals are analyzed to obtain the tooth surface etching potential value, including:

[0075] The timing characteristics of torque and speed signal changes during the target process are acquired. These timing characteristics include: sequence, duration, and rhythm of change, among other time-related features. Based on these timing characteristics, the target process is divided into three stages: startup, ramp-up, and peak maintenance. The duration of each stage is then calculated. Specifically, this includes: acquiring the historical maximum torque of the target object during the target process; capturing signal changes with a 10ms sliding window step; and capturing the torque as it rises from its initial value (zero-speed torque) and the speed increases from 0 to the rated speed (for the pontoon winch motor under rated voltage, rated frequency, and rated load). When the rotor speed (i.e., the rated speed) reaches 20% of the rated speed under the operating conditions, it is marked as the start-up phase, which lasts until the torque reaches 30% of the peak value (historical maximum torque). The period when the torque rises from 30% of the peak value to the peak value and the speed increases from 20% of the rated speed to the rated speed is the climbing phase, which ends when the torque first reaches the peak value. The period when the torque is stable within ±5% of the peak value and the speed is stable within ±3% of the rated speed is the peak phase (peak value maintenance). Dividing the duration of each phase by the total duration of the target process gives the percentage of the duration of each phase: start-up, climbing, and peak.

[0076] Based on the phase duration proportion, the preprocessed torque and speed signals are enhanced in stages to generate phased torque and speed feature quantities. Specifically, this includes: In the start-up phase, calculating the torque rise rate (calculated by the average difference between torque sampling points within adjacent sliding windows) and the speed increase rate (calculated by the average difference between speed sampling points within adjacent sliding windows), and multiplying the torque rise rate and speed increase rate by the phase duration proportion of the start-up phase to obtain the torque feature weight and speed feature weight of the start-up phase; In the climb phase, dividing the torque difference between the first and last sampling points within the phase by the total phase duration to obtain the average torque change rate; dividing the speed difference between the first and last sampling points by the total phase duration to obtain the average speed change rate, and multiplying the average torque change rate by the phase duration proportion of the climb phase to obtain the torque trend of the climb phase. Weighting: Multiply the average rate of change of rotational speed by the proportion of the climbing phase duration to obtain the rotational speed trend weight of the climbing phase; calculate the difference between the torque value of each sliding window within the peak phase and the peak torque, square all differences, take the arithmetic mean, and then take the square root to obtain the torque fluctuation value; calculate the difference between the rotational speed value of each window and the rated speed, square all differences, take the arithmetic mean, and then take the square root to obtain the rotational speed fluctuation value; multiply the torque fluctuation value by the proportion of the peak phase duration to obtain the peak phase torque stability weight; multiply the rotational speed fluctuation value by the proportion of the peak phase duration to obtain the peak phase rotational speed stability weight; add the starting phase torque characteristic weight, starting phase rotational speed characteristic weight, climbing phase torque trend weight, climbing phase rotational speed trend weight, peak phase torque stability weight, and peak phase rotational speed stability weight together to obtain the phased torque and rotational speed characteristic quantities;

[0077] Based on the phased torque and speed characteristics, the correlation between torque and speed in the target process is analyzed to generate a torque-speed linkage coefficient. Specifically, this includes: for the starting phase, multiplying the starting torque characteristic weight by the starting speed characteristic weight and then by the phase duration proportion of the starting phase to obtain the starting linkage value; multiplying the product of the torque trend weight and speed trend weight in the climbing phase by the phase duration proportion of the climbing phase to obtain the climbing linkage value; multiplying the product of the torque stability weight and speed stability weight in the peak phase by the phase duration proportion of the peak phase to obtain the peak linkage value; and adding the starting linkage value, climbing linkage value, and peak linkage value to obtain the torque-speed linkage coefficient.

[0078] Obtain the structural properties of the target object, combine the torque-speed linkage coefficient with the structural properties to calculate the tooth surface stress fluctuation value. The structural properties include: number of teeth, tooth profile, module, tooth width, material hardness, etc.

[0079] The preprocessed torque and speed signals are calculated to obtain the rates of change of torque and speed. Specifically, this includes: dividing the torque difference between the current window and the previous window by the window time interval (10ms) with a window step size of 10ms to obtain the instantaneous rate of change of torque; dividing the speed difference between the current window and the previous window by the window time interval (10ms) to obtain the instantaneous rate of change of speed; and calculating the average of the instantaneous rates of change of torque and speed for all windows in each stage to obtain the rate of change of torque and speed for the corresponding stage.

[0080] The tooth surface stress fluctuation value and its rate of change are combined to obtain the tooth surface erosion potential value. Specifically, this involves: multiplying the stress fluctuation value by the torque change rate, then by the rotational speed change rate, and then by the proportion of the start-up phase duration to obtain the start-up phase erosion contribution value; multiplying the stress fluctuation value by the torque change rate, then by the rotational speed change rate, and then by the proportion of the climb phase duration to obtain the climb phase erosion contribution value; multiplying the stress fluctuation value by the torque change rate, then by the rotational speed change rate, and then by the proportion of the peak phase duration to obtain the peak phase erosion contribution value; and finally, adding the start-up phase erosion contribution value, the climb phase erosion contribution value, and the peak phase erosion contribution value to obtain the tooth surface erosion potential value.

[0081] Signal analysis improves the accuracy of identifying early latent damage to barge winch gears. It divides the zero-speed-peak torque switching process into three stages: start-up, ramp-up, and peak maintenance. Combining a 10ms sliding window to capture timing features, it calculates torque and speed characteristics by weighting the proportion of stage duration. This allows for precise quantification of the dynamic impact intensity at different stages. In particular, by combining gear structural properties with the torque-speed linkage coefficient to generate tooth surface stress fluctuation values, and integrating torque and speed change rates to obtain corrosion potential values, it can capture early signs of latent damage such as minor tooth surface scuffing and abrasion.

[0082] By calculating the corrosion contribution value in stages, the influence weight of different working conditions on gear damage can be dynamically reflected, making the tooth surface corrosion potential value more consistent with the actual working conditions of frequent start-stop of barge winches. This allows for a dynamic match between gear oil replacement and the progression of hidden gear damage, effectively solving the problem of mismatch between the protective effect of new oil and actual damage, and reducing the accumulation of hidden wear.

[0083] In one embodiment, the torque-speed linkage coefficient and structural properties are combined for calculation to obtain the tooth surface stress fluctuation value, including:

[0084] The structural properties are analyzed, and weights are assigned according to the stress sensitivity of the target object in the target process to generate structural dynamic characteristic values. Specifically, the weights are set according to the stress sensitivity: module weight is 0.3, tooth width weight is 0.25, material hardness weight is 0.2, tooth profile weight is 0.15, and tooth count weight is 0.1, with a total weight of 1. The actual value of each attribute is divided by its maximum value (the maximum value allowed in industry design standards, such as the maximum module of the winch gear of the same specification), and then multiplied by the corresponding weight. The multiple products are then added together to obtain the structural dynamic characteristic values.

[0085] The variation law of structural dynamic characteristic value and torque-speed linkage coefficient in each stage is analyzed to obtain the structure-linkage dynamic coupling coefficient. Specifically, the structure dynamic characteristic value is multiplied by the torque-speed linkage coefficient, and then multiplied by the duration ratio of the start-up, climb and peak stages respectively to obtain the coupling components of the start-up, climb and peak stages. The coupling components of the three stages are added together to obtain the structure-linkage dynamic coupling coefficient.

[0086] Based on structural properties, the deformation at different meshing positions of the gear during contact is calculated to obtain the gear contact deformation. Specifically, this includes: dividing the actual hardness of the gear by the standard hardness, multiplying by the gear module, and then dividing by the tooth width to obtain the tooth tip deformation; calculating the tooth root thickness based on the number of teeth and the module (number of teeth × module × 0.8), multiplying the tooth root thickness by the material hardness ratio (actual hardness of the gear divided by the standard hardness), and then dividing by the tooth width to obtain the tooth root deformation; dividing the product of the module and the tooth width by the material hardness ratio to obtain the tooth surface mid-section deformation; and adding the tooth tip deformation, tooth root deformation, and tooth surface mid-section deformation to obtain the gear contact deformation.

[0087] The structure-linkage dynamic coupling coefficient is calibrated based on the gear contact deformation to generate the tooth surface stress fluctuation value. Specifically, the structure-linkage dynamic coupling coefficient is multiplied by the gear contact deformation to obtain the preliminary stress value, and the material hardness ratio is multiplied by the preliminary stress value to obtain the tooth surface stress fluctuation value.

[0088] By integrating the torque-speed linkage coefficient with the gear structural properties, and assigning weights to structural properties such as module and tooth width according to the degree of stress sensitivity, and generating structural dynamic characteristic values, the structure-linkage dynamic coupling coefficient is obtained by combining the duration ratio of each stage. Then, by calibrating the gear contact deformation and introducing the material hardness ratio, the stress fluctuation value of the tooth surface can accurately map the stress deformation difference at different meshing positions, thereby understanding the subtle stress changes under the frequent switching conditions of the barge winch, and making maintenance decisions more in line with the actual damage state of the gear.

[0089] In one embodiment, the tooth surface alteration potential is calculated based on the tooth surface temperature and vibration frequency to obtain the alteration conductivity, including:

[0090] Based on the tooth surface temperature, the frequency and duration of temperature surge points during the target process are analyzed to generate thermal shock characteristic values. Specifically, this includes: setting a temperature surge threshold of 10℃ / 10ms, counting the number of times the threshold is exceeded during the target process (frequency), multiplying the window number of each surge by 10 to obtain the total duration; multiplying the frequency by 0.6 (weight) and the total duration by 0.4 (weight) to obtain the thermal shock characteristic values. Among them, during the zero-speed to peak torque switching process of the barge winch gear disk, the winch gear is subjected to instantaneous impact loads during the switching process. The risk of instantaneous lubrication failure of gear oil (such as sudden rupture of the oil film) has a more direct impact on tooth surface damage. The frequency reflects the degree of instantaneous failure. The higher the frequency, the more violent the fluctuation of lubrication capacity and the greater the risk of instantaneous damage. Therefore, it is given a higher weight (0.6). The total duration reflects the cumulative effect of lubrication failure. Although it will also aggravate tooth surface wear, compared with the high frequency of instantaneous failure, its impact on the instantaneous damage of the tooth surface during the current torque switching process is smaller. Therefore, it has a lower weight (0.4).

[0091] Based on the vibration frequency, the jump amplitude and duration period of abnormal vibration during the target process are extracted to obtain the vibration frequency anomaly feature value. Specifically, this includes: setting the abnormal vibration threshold to ±15% of the rated frequency (rated operating frequency of the gear disk); counting the jump amplitude exceeding the threshold (the difference between the current window frequency and the previous window frequency is the jump amplitude); summing the jump amplitudes of all windows to obtain the total jump amplitude; counting the abnormal duration period exceeding the threshold each time (number of abnormal duration windows × 10); summing all abnormal duration periods to obtain the total period; multiplying the total jump amplitude by 0.7 and the total period by 0.3 to obtain the vibration frequency anomaly feature value.

[0092] The micro-texture structure parameters of the gear tooth surface are obtained. Based on these parameters, the thermal shock characteristic value and the vibration frequency anomaly characteristic value are calculated to obtain the thermal shock index and the vibration frequency anomaly index. The micro-texture structure parameters include tooth surface roughness, texture direction angle, texture depth, etc. Specifically, the thermal shock index is obtained by multiplying the tooth surface roughness by the thermal shock characteristic value, adding the texture depth by the thermal shock characteristic value, and then multiplying by 0.5. The vibration frequency anomaly index is obtained by adding the product of the texture direction angle and the vibration frequency anomaly characteristic value to the product of the tooth surface roughness and the vibration frequency anomaly characteristic value, and then multiplying by 0.5.

[0093] The synchronicity of the thermal shock index and the frequency anomaly index on the time axis is analyzed to generate a thermal-vibration synergy factor. Specifically, this involves: using a fixed window of 10ms, traversing the time axis of the entire target process, with each window corresponding to a set of thermal shock index and frequency anomaly index; when the thermal shock index > 0.6 and the frequency anomaly index > 0.5 within a certain window, it is marked as a synchronized window; the number of all synchronized windows is divided by the total number of windows to obtain the synchronization rate; for each synchronized window, the product of the thermal shock index and the frequency anomaly index within that window is calculated, and the product results of all synchronized windows are summed and divided by the number of synchronized windows to obtain the average product; the average product is multiplied by the synchronization rate to obtain the thermal-vibration synergy factor.

[0094] By analyzing tooth surface temperature and vibration frequency, and considering the frequent switching conditions of barge winches, the frequency and duration of temperature surges are statistically analyzed with a threshold of 10℃ / 10ms. Combined with weights, thermal shock characteristic values ​​are calculated to understand the risk of immediate lubrication failure of gear oil. At the same time, by extracting the jump amplitude and period through abnormal vibration thresholds, vibration frequency anomaly characteristic values ​​are generated, which can keenly detect abnormal gear meshing and help to intervene in lubrication failure and meshing abnormality problems in advance.

[0095] By combining the synergistic analysis of gear microstructure parameters and thermal vibration characteristics, and calibrating thermal shock and vibration frequency anomaly characteristics using parameters such as tooth surface roughness and texture depth, thermal shock index and vibration frequency anomaly index are generated. The synchronicity of the two on the time axis is calculated to obtain the thermal vibration synergy factor, which can accurately map the formation process of hidden damage such as minor scuffing and scratches. This makes the calculation of alteration conductivity more consistent with the actual damage to gears, avoids the neglect of hidden damage in traditional maintenance, and makes maintenance methods more targeted.

[0096] In one embodiment, the tooth surface alteration potential is calculated based on the tooth surface temperature and vibration frequency to obtain the alteration conductivity, and the method further includes:

[0097] The mapping relationship between the thermal vibration synergy factor and the tooth surface alteration potential value is analyzed to obtain the synergy-alteration correlation degree. Specifically, this includes: calculating the absolute value of the difference between the thermal vibration synergy factor and the tooth surface alteration potential value within each 10ms window, summing the absolute values ​​of the differences of all windows to obtain the total deviation; calculating the total number of windows in the target process multiplied by the product of the thermal vibration synergy factor and the tooth surface alteration potential value to obtain the baseline value; and subtracting (total deviation divided by baseline value) from 1 to obtain the synergy-alteration correlation degree.

[0098] The influence of different tooth surface temperatures and vibration frequencies on the gear is calculated. Based on the influence values, the synergistic-alteration correlation degree is corrected to obtain the alteration conductivity. Specifically, this involves: multiplying (real-time tooth surface temperature minus standard tooth surface temperature and then dividing by standard tooth surface temperature) by 0.005 to obtain the fatigue strength attenuation rate at different temperatures; multiplying (actual vibration frequency divided by standard vibration frequency and then minus 1) by 0.003 to obtain the vibration frequency influence coefficient; multiplying the fatigue strength attenuation rate and the vibration frequency influence coefficient by 0.5 respectively and then adding them together to obtain the total influence value; and multiplying the synergistic-alteration correlation degree by (1 + total influence value) to obtain the alteration conductivity.

[0099] By constructing a mapping relationship between the thermal vibration synergy factor and the tooth surface etching potential value, and combining the actual influence of temperature and vibration for correction, the accuracy of etching conductivity is greatly improved. The total deviation and the benchmark value are calculated in 10ms windows. The resulting synergy-etching correlation degree can accurately reflect the intrinsic connection between thermal vibration synergy and tooth surface etching. The total influence value is then calculated by the fatigue strength attenuation rate and the vibration frequency influence coefficient to correct the correlation degree. This allows the etching conductivity to dynamically fit the actual damage state of the gear, enabling an understanding of the damage situation under frequent switching conditions of the barge winch, which facilitates later maintenance and upkeep.

[0100] In one embodiment, the preprocessed image data is analyzed to generate a lubrication attenuation coefficient, including:

[0101] Image data is analyzed to capture the discrete trajectories and aggregation frequencies of microstructures within the oil, generating an oil phase dispersion indicator. Specifically, this involves: capturing preprocessed image data in 10ms windows; setting grayscale thresholds (above 200 for bright areas, below 50 for dark areas); scanning the pixel grayscale values ​​of the oil image within the 10ms window; identifying continuous edge contours with abrupt grayscale changes; and determining microstructures when the contour is closed and its area is between 5 and 500 pixels. Dark areas with closed contours are marked as contaminant particles, and bright areas with irregular edges are marked as oil film fragments. Each microstructure's pixels are also labeled. The movement distance of the same microstructure in adjacent windows is calculated. The distance is calculated by squarening the pixel differences in the horizontal and vertical directions, summing the results, and then taking the square root of the sum to obtain the straight-line distance in pixels. This straight-line distance is then multiplied by the actual length of a single pixel (0.01 mm) to obtain the actual straight-line distance between two points, which is the moving distance. The average moving distance of all microstructures is used as the average trajectory length within the window. The number of times the distance between any two microstructures within the window is less than 5 pixels is counted and used as the aggregation frequency. The average trajectory length is multiplied by 0.6, and the aggregation frequency is multiplied by 0.4 to obtain the dispersion of the window. The average dispersion of all windows is then calculated to obtain the oil phase dispersion indicator.

[0102] The calculation of the light and shadow gradient changes of the oil film during gear meshing in image data yields the stable value of the oil film boundary. Specifically, this involves: setting a grayscale gradient threshold of 30-150 for the image data; filtering pixel regions with gradient values ​​greater than the threshold; connecting continuous closed contours to form a continuous region that is the oil film candidate area; setting the tooth tip feature as a convex arc with a radius of curvature of 5-8 pixels; setting the tooth root feature as a concave arc with a radius of curvature of 3-5 pixels; traversing the oil film candidate area to match a complete gear contour line containing tooth tip and tooth root features; marking the tip point (tooth tip) of each tooth and the bottom point (tooth root) between two teeth; and connecting the vertices of adjacent gears to form a tooth tip connection line. Connect the tooth root points on the same side to form a tooth root line; the area enclosed by the intersection of the tooth tip line and the tooth root line is the meshing area; extract the image of the intersection area of ​​the oil film candidate area and the meshing area as the meshing area oil film image within a 10ms window; calculate the gradient value of the light and shadow of each pixel in the meshing area oil film image (obtain the square value of the horizontal gradient and the square value of the vertical gradient respectively, add the two square values, and then take the square root of the sum to get the gradient value); divide the number of pixels with gradient values ​​< 20 by the total number of pixels in the oil film image to obtain the stable value of the window; calculate the arithmetic mean of the stable values ​​of all windows to obtain the oil film boundary stable value.

[0103] The correlation between oil phase dispersion indicators and oil film boundary stability values ​​during gear operation cycles was analyzed to generate performance degradation coupling values. Specifically, this involved: using gear operation cycles as units (e.g., 30 10ms windows), multiplying the oil phase dispersion indicator by 0.4 and the oil film boundary stability value by 0.6 for each window to obtain the coupling value for that window; calculating the mean of all window coupling values ​​yielded the performance degradation coupling value. For barge winch gears, the oil phase dispersion indicator can reflect the microstructure within the gear oil (contaminants, etc.). The dispersion state of oil film fragments mainly reflects the uniformity of the oil and is a basic condition for lubrication efficiency, but it is an indirect influence, so its weight is low at 0.4. The oil film boundary stability value directly reflects the stability of the oil film in the gear meshing area (such as whether it breaks or whether it continuously covers the tooth surface). When the barge winch gear is subjected to instantaneous impact during the zero speed to peak torque switching, the stability of the oil film is the core guarantee to prevent tooth surface wear and erosion, and directly determines the lubrication effectiveness. Therefore, its weight is high at 0.6, which is more in line with the core requirements of lubrication protection under instantaneous impact.

[0104] The analysis of image data reveals the magnitude and speed of oil morphological changes with torque and rotational speed, generating a working condition adaptation variation rate. Specifically, this involves: traversing each pixel of the oil film candidate area within a 10ms window, defining the actual area of ​​each pixel (1 pixel = 0.01mm²), multiplying the total number of pixels in the oil film candidate area by the area of ​​a single pixel to obtain the oil film area enclosed by the contour (oil film contour); dividing the difference in oil film area between the current window and the previous window by the initial window oil film area (when torque is zero) to obtain the morphological change magnitude; dividing the morphological change magnitude by the window interval to obtain the morphological change speed; multiplying the morphological change magnitude by 0.6 and the morphological change speed by 0.4 to obtain the variation component of each window; and calculating the arithmetic mean of all window variation components to obtain the working condition adaptation variation rate.

[0105] The efficiency degradation coupling value and the operating condition adaptation variation rate are fused to obtain the efficiency degradation coefficient. Specifically, for each 10ms window, the efficiency degradation coupling value is multiplied by the operating condition adaptation variation rate and then multiplied by 0.5 to obtain the synergistic effect parameter; the absolute value of the difference between the efficiency degradation coupling value and the operating condition adaptation variation rate is multiplied by 0.5 to obtain the antagonistic parameter; the synergistic effect parameter and the antagonistic parameter are added together to obtain the fused value, and the mean of the fused values ​​for all windows is calculated, which is the efficiency degradation coefficient.

[0106] By analyzing gear oil image data from multiple dimensions, the discrete trajectory and aggregation frequency of the oil microstructure are captured within a 10ms window. Combined with grayscale thresholding to identify contaminants and oil film fragments, the generated oil phase dispersion index reflects the uniformity of the oil. Simultaneously, the stability value of the oil film boundary in the meshing zone is calculated through changes in light and shadow gradients, directly reflecting the stability of the oil film and capturing early failure signs of gear oil under frequent impacts. This facilitates later judgment on whether gear oil replacement is necessary. Furthermore, by fusing the efficiency attenuation coupling value and the operating condition adaptation variation rate, a lubrication attenuation coefficient is generated, achieving dynamic matching between gear oil performance and operating conditions. Considering the instantaneous impact characteristics of barge winches, a higher weight is given to the oil film stability value. Combined with the amplitude and speed of changes in oil morphology with torque and speed, the lubrication attenuation coefficient accurately maps the real-time changes in gear oil protection capability, solving the mismatch between newly replaced gear oil and the progression of hidden gear damage. This makes maintenance measures more aligned with actual needs, reduces the accumulation of hidden wear, and significantly improves the maintenance effect of barge winches.

[0107] In one embodiment, the alteration conductivity and the lubrication attenuation coefficient are matched to obtain the maintenance imbalance index, including:

[0108] The alteration conductivity and lubrication attenuation coefficient are expanded in terms of feature dimensions to generate alteration feature vectors and lubrication feature vectors, respectively. Specifically, this includes: calculating the average alteration conductivity at each stage of startup, ramp-up, and peak; traversing all windows of the entire target process to find the maximum and minimum values ​​of alteration conductivity, and arranging the five features of the average, maximum, and minimum alteration conductivity at each stage of startup, ramp-up, and peak in order to form the alteration feature vector; calculating the average lubrication attenuation coefficient for each gear operation cycle to obtain the cycle average; calculating the difference between the current cycle average and the previous cycle average to obtain the cycle fluctuation value; traversing all windows of the entire process to find the maximum and minimum values ​​of the lubrication attenuation coefficient; and arranging the four features of the cycle average, cycle fluctuation value, maximum value, and minimum value in order to form the lubrication feature vector.

[0109] Image data is analyzed to obtain gear oil viscosity. This viscosity is then combined with microstructure parameters to obtain a viscosity-roughness matching value. Specifically, this involves: marking the center coordinates of the oil film profile in 10ms windows; calculating the straight-line distance between the center coordinates of the current and previous windows; and dividing this distance by 10 to obtain the average flow velocity of the oil film. The viscosity corresponding to each velocity is set as follows: 0.2 mm / ms corresponds to a viscosity of 150 cSt, 0.3 mm / ms corresponds to 120 cSt (with a decrease of 30 cSt for every 0.1 mm / ms interval), thus enabling... The viscosity is obtained by measuring the average flow velocity of the oil film. The viscosity is multiplied by 0.6, and the tooth surface roughness is multiplied by 0.4. The sum of these two values ​​is the viscosity-roughness matching value. For the gears of the barge winch, the gear oil viscosity directly determines the oil film load-bearing capacity. Under the instantaneous impact of switching from zero speed to peak torque, the gear oil viscosity needs to match the impact load to maintain the integrity of the oil film and plays a dominant role in lubrication effectiveness. Therefore, it has a higher weight of 0.6. On the other hand, the tooth surface roughness reflects the microscopic unevenness of the tooth surface and affects the oil film adhesion stability, but it plays a more auxiliary role. Therefore, it has a lower weight of 0.4.

[0110] The spatial matching degree between the alteration feature vector and the lubrication feature vector is calculated to obtain the alteration-lubrication matching deviation value. Specifically, this includes: since the alteration feature vector contains 5 elements (starting mean, rising mean, peak mean, maximum value, minimum value) and the lubrication feature vector contains 4 elements (period mean, period fluctuation value, maximum value, minimum value), the last element of the lubrication feature vector is padded with 0 to form a 5-element vector, ensuring that the two vectors have the same dimension. The difference between the corresponding elements of the alteration feature vector and the lubrication feature vector is calculated, and the square root of the sum of the squares of each difference is obtained to obtain the Euclidean distance. The magnitude of the alteration feature vector and the lubrication feature vector (the magnitude is the square root of the sum of the squares of each element) is calculated separately, and the two magnitudes are added to obtain the total magnitude. The Euclidean distance is divided by the total magnitude to obtain the alteration-lubrication matching deviation value.

[0111] The lubrication attenuation rate is calculated based on the alteration conductivity as a function of the gear tooth surface alteration degree. Specifically, this involves: using 10ms as a window, calculating the difference in alteration conductivity between the current window and the previous window in each of the startup, ramp-up, and peak stages, and dividing the difference by 10 to obtain the instantaneous attenuation rate; calculating the average instantaneous rate of all windows in each stage to obtain the stage average attenuation rate; and multiplying the three stage average rates by the corresponding stage duration percentage and summing them to obtain the lubrication attenuation rate value.

[0112] The viscosity-roughness fit value and lubrication decay rate value are used as weighting parameters to adjust the alteration-lubrication energy matching deviation value, generating a dynamic fit imbalance. Specifically, this involves: multiplying the viscosity-roughness fit value by the lubrication decay rate value to obtain the coupling coefficient; multiplying the alteration-lubrication energy matching deviation value by both the viscosity-roughness fit value and the lubrication decay rate value, then adding the two products and dividing by (viscosity-roughness fit value plus lubrication decay rate value) to obtain the baseline imbalance; and multiplying the coupling coefficient by the baseline imbalance to obtain the dynamic fit imbalance.

[0113] Based on the dynamic adaptation imbalance, the peak ranges of torque and speed signals are mapped to generate a maintenance imbalance index.

[0114] By expanding the alteration conductivity into a feature vector based on the stage mean and extreme values, and expanding the lubrication attenuation coefficient into a vector based on the periodic characteristics, and calculating the spatial matching degree after zeroing to unify the dimension, the obtained alteration-lubrication matching deviation value can intuitively reflect the adaptation difference between the two. At the same time, combined with the adaptation value of gear oil viscosity and tooth surface roughness, as well as the lubrication effect attenuation rate with alteration, the deviation value is weighted and adjusted so that the dynamic adaptation imbalance can fit the actual working conditions under the instantaneous impact of the barge winch, breaking through the limitations of the traditional maintenance method of separating the gear and oil state assessment.

[0115] By dynamically adapting the imbalance degree to map the peak torque-speed range to generate a maintenance imbalance index, a synergistic assessment of damage and lubrication is achieved. This allows for the detection of hidden damage such as minor adhesion and also reflects the decline in the protective capability of gear oil. This effectively solves the problem of mismatch between new oil changes and actual damage, and improves the long-term stability maintenance of equipment.

[0116] In one embodiment, based on the dynamic adaptation imbalance, the peak ranges of the torque signal and the speed signal are mapped to generate a maintenance imbalance index, including:

[0117] The transient abrupt change characteristics (including signal rise edge steepness, peak drop rate, and abnormal pulse density) within the peak range of torque and speed signals are analyzed to generate transient peak change characteristic values. Specifically, this includes: dividing the torque and speed peak ranges (times when torque is stable within ±5% of the peak value and speed is stable within ±3% of the rated speed) into continuous windows at 10ms intervals, calculating the torque difference (speed difference) between each window and the previous window, and dividing the torque difference by 10 to obtain the instantaneous steepness; traversing all windows in the peak range, selecting the largest instantaneous steepness as the signal rise edge steepness; after the peak range ends, calculating the absolute value of the torque (speed) difference between the first window and the peak window, and dividing the absolute value by 10 to obtain the drop rate; dividing the number of windows in the peak range where the torque (speed) exceeds ±10% of the peak value by the total number of windows in the range to obtain the proportion; multiplying the signal rise edge steepness by 0.35, the drop rate by 0.35, and the proportion by 0.3, and then adding the three products together to obtain the transient peak change characteristic value.

[0118] The interaction between transient peak variation eigenvalues ​​and dynamic adaptation imbalance is analyzed at each stage of the target process to obtain the spatiotemporal coupling imbalance coefficient. Specifically, in the three stages of initiation, ramp-up, and peak, the product of transient peak variation eigenvalues ​​and dynamic adaptation imbalance and the absolute value of the difference are calculated in each stage to obtain the synergistic and antagonistic terms. The synergistic and antagonistic terms are multiplied by 0.5 and then added to obtain the stage coupling value. The spatiotemporal coupling imbalance coefficient is obtained by multiplying the three stage coupling values ​​by the corresponding stage duration proportions and then adding them.

[0119] Based on image data, the critical value for oil film rupture during gear meshing is calculated. Specifically, this involves: establishing a correspondence between grayscale values ​​and actual oil film thickness in the image, using a 10ms window: a grayscale value of 200 corresponds to a thickness of 0.5μm, and the thickness increases by 0.1μm for every 10 decrease in grayscale value; traversing the oil film image of the meshing area, marking windows where the oil film contour shows a break (continuous edge interruption exceeding 5 pixels) as rupture windows, and finding the three consecutive normal windows preceding the rupture window (i.e., if the rupture window is the nth, then it corresponds to the (n-1), (n-2), and (n-3)th normal windows). -3 windows); For these 3 normal windows, calculate the minimum thickness of the oil film in each window (each window has only 1 minimum value; if multiple pixels in the same window have the same thickness and are all minimum values, they are still counted as 1 value); if there are multiple broken windows, extract the minimum thickness of 3 normal windows for each broken window; add up all the extracted minimum thickness values ​​and divide by the total number of extracted minimum thickness values ​​(i.e., each broken window corresponds to 3 values, and the total number is the number of broken windows multiplied by 3) to obtain the critical value for oil film rupture;

[0120] The spatiotemporal coupling imbalance coefficient is dynamically calibrated based on the oil film rupture critical value to generate the calibrated spatiotemporal imbalance value. Specifically, this involves: dividing the actual thickness of the oil film within the current 10ms window by the oil film rupture critical value to obtain the critical approach degree; multiplying the spatiotemporal coupling imbalance coefficient by (1 + critical approach degree multiplied by 0.5) to obtain the calibration coefficient; and then multiplying the spatiotemporal coupling imbalance coefficient by the calibration coefficient to obtain the calibrated spatiotemporal imbalance value.

[0121] The spatiotemporal imbalance value is converted to generate a maintenance imbalance index. Specifically, this involves finding the maximum value of the spatiotemporal imbalance value after calibration during the target process, dividing the spatiotemporal imbalance value by the maximum value to obtain a normalization coefficient, and multiplying the normalization coefficient by 100 and taking the integer part to obtain the maintenance imbalance index (range 0-100, the higher the value, the more severe the imbalance).

[0122] By analyzing and generating a maintenance imbalance index, we can understand the degree of imbalance between gear damage and lubrication protection in barge winches. By analyzing the transient change characteristics of torque and speed peak ranges, and combining the dynamic adaptation imbalance degree, we obtain the spatiotemporal coupling imbalance coefficient. Then, based on the oil film rupture critical value, we finally generate a maintenance imbalance index of 0-100. This index not only captures the impact of signal changes on gears, but also incorporates the key influence of oil film stability. It can keenly reflect the correlation between hidden damage such as slight adhesion and lubrication failure, avoid the problem of mismatch between new oil and damage, reduce the accumulation of hidden wear, and improve maintenance effectiveness.

[0123] In one embodiment, the maintenance imbalance index is calculated based on the service life of the gears and gear oil to obtain a comprehensive maintenance value, including:

[0124] This study analyzes the performance degradation of gears and gear oil across different service life intervals, generating nonlinear degradation coefficients for both gears and gear oil. Specifically, it involves: obtaining the service life of the gears and gear oil, dividing the service life into three intervals: 0-30%, 30%-70%, and 70%-100%, with weights of 0.2, 0.5, and 0.3 respectively; using a new gear tooth surface image as a baseline (grayscale value set to 200), analyzing the current tooth surface image within a 10ms window, and marking continuous pixel areas with grayscale values ​​below 20% of the baseline (i.e., <160) as wear zones; the actual area corresponding to a single pixel is... 0.01mm², multiply the total number of pixels in the wear area by 0.01mm² to obtain the wear area area; calculate the difference between the grayscale value of each pixel in the wear area and the reference grayscale (reference grayscale is 200), and increase the depth by 0.01mm for every 10 differences. Multiply the difference by 10 and then by 0.01mm to obtain the wear pixel depth. Calculate the average depth of all wear pixel depths; multiply the wear area by the average depth to obtain the actual wear amount of the gear; set the total limit according to the gear material hardness (HB) = hardness value × 0.02mm (for example, the total limit for a 300HB gear = 6mm); where 0- The maximum permissible wear for the 30% lifespan range is 20% of the total limit; for the 30%-70% lifespan range, it is 50%; and for the 70%-100% lifespan range, it is 100%. For each range, the actual wear is divided by the maximum permissible wear for that range. The three products are then multiplied by their respective range weights of 0.2, 0.5, and 0.3, and summed to obtain the gear aging nonlinear decay coefficient. The initial viscosity of the gear oil is set to the factory viscosity. The average oil film flow velocity is calculated as follows: 0.2 mm / ms corresponds to 150 cSt, and for every 0.1 mm / ms increase... The viscosity is calculated by subtracting 30 cSt from the initial viscosity (ms viscosity) to obtain the current viscosity. The actual viscosity decay rate is obtained by subtracting the current viscosity from the initial viscosity and then dividing by the initial viscosity. The maximum decay rate is 10% of the initial viscosity in the 0-30% lifespan range, 20% in the 30%-70% lifespan range, and 30% in the 70%-100% lifespan range. For each range, the actual decay rate is divided by the maximum decay rate of that range. The product of the three ranges is then multiplied by the corresponding range weights of 0.2, 0.5, and 0.3, respectively, and summed to obtain the nonlinear decay coefficient of the oil aging process.

[0125] The dynamic change rates of the gear aging nonlinear decay coefficient and the oil aging nonlinear decay coefficient are calculated to obtain the duration-decay coordination coefficient. Specifically, this includes: taking the gear aging nonlinear decay coefficient for three consecutive windows of 10ms, dividing the difference between the previous and subsequent windows by 20 to obtain the gear average rate; taking the oil aging nonlinear decay coefficient for three consecutive windows of 10ms, dividing the difference between the previous and subsequent windows by 20 to obtain the oil average rate; multiplying the gear average rates of the three windows by weights of 0.3, 0.5, and 0.2 respectively, and then summing them to obtain the gear total weighted rate; multiplying the oil average rates of the three windows by weights of 0.3, 0.5, and 0.2 respectively, and then summing them to obtain the oil total weighted rate. Weighted rate; multiplying the total weighted rate of gears by 0.4 and adding the total weighted rate of oil by 0.6 yields the duration-degradation synergy coefficient; among them, during the zero-speed-peak torque switching process of the pontoon winch, the lubrication performance of gear oil is the core guarantee to prevent instantaneous impact damage to the tooth surface, while the performance degradation of oil (gear oil) (such as decreased viscosity and reduced oil film stability) will directly lead to lubrication failure. Under instantaneous high torque impact, oil film rupture may instantly aggravate tooth surface wear and erosion. Its dynamic changes are more sensitive to the timing of maintenance, so the weight of the total weighted rate of oil is relatively high at 0.6; while gear wear is the result of long-term accumulation, and its degradation rate has a relatively lagging impact on immediate maintenance decisions, so the weight of the total weighted rate of gears is relatively low at 0.4;

[0126] The maintenance imbalance index is adjusted in stages based on the duration-degradation synergy coefficient to generate a dynamic maintenance estimate. Specifically, this includes: multiplying the maintenance imbalance index by the duration-degradation synergy coefficient in each of the 0-30%, 30%-70%, and 70%-100% lifespan ranges to obtain the stage adjustment value; and then multiplying the three stage adjustment values ​​by weights of 0.2, 0.5, and 0.3 respectively and summing them to obtain the dynamic maintenance estimate.

[0127] A multidimensional attenuation factor is generated by fusing the tooth surface stress fluctuation value, thermal shock characteristic value, vibration frequency anomaly index, and oil film rupture critical value. Specifically, this involves: traversing the entire target process in 10ms windows; in each window, multiplying the tooth surface stress fluctuation value by the vibration frequency anomaly index to obtain the mechanical shock component; multiplying the thermal shock characteristic value by the oil film rupture critical value to obtain the lubrication failure component; multiplying the mechanical shock component by 0.55 and the lubrication failure component by 0.45 and then adding them together to obtain the window factor; and calculating the mean of all window factors to obtain the multidimensional attenuation factor.

[0128] The dynamic maintenance estimate is corrected for deviations based on a multidimensional decay factor to obtain the comprehensive maintenance value. Specifically, this involves: multiplying the dynamic maintenance estimate by the multidimensional decay factor to obtain a basic correction value; multiplying the absolute value of the difference between the dynamic maintenance estimate and the multidimensional decay factor by 0.3, and then adding this to the basic correction value to obtain the comprehensive maintenance value (if the comprehensive maintenance value...). If the value is 0, then the value is 0. If the comprehensive maintenance value... If the value is 100, then the value is 100; if it is 0... Comprehensive maintenance value If the value is 100, the original value is retained. The range of the comprehensive maintenance value is between 0 and 100.

[0129] By dividing the service life into three intervals and assigning different weights, and calculating the nonlinear decay coefficient by combining the actual wear amount and viscosity decay rate, and then obtaining the time-decay coordination coefficient through the dynamic change rate, this method can better match the nonlinear characteristics of performance decay during long-term use of barge winches. This makes the adjustment of the maintenance imbalance index more consistent with the actual wear pattern. In addition, by correcting the dynamic maintenance estimate through multi-dimensional decay factors, a comprehensive maintenance value is generated. Furthermore, by integrating mechanical and lubrication factors such as tooth surface stress fluctuations and thermal shock, the deviation of the maintenance estimate at different life stages is corrected. This allows the comprehensive maintenance value to accurately reflect the dynamic matching state between hidden gear damage and oil protection capability, solving the problem of mismatch between new oil changes and actual damage progress, and significantly improving the effectiveness of barge winch maintenance.

[0130] In one embodiment, determining a maintenance instruction based on the comprehensive maintenance value includes:

[0131] Based on the comprehensive maintenance value, maintenance intervals and maintenance instructions are determined, specifically including: setting four maintenance intervals, namely:

[0132] The first nutrient range is 61-80;

[0133] The second nutrient range is: 81-100;

[0134] The third nutrient range is: 41-60;

[0135] The fourth dimensional range is 0-40;

[0136] When the overall maintenance value is in the first maintenance range, the first maintenance instruction is triggered: replace the gear oil separately;

[0137] When the overall maintenance value is in the second maintenance range, the second maintenance instruction is triggered: replace the gear individually;

[0138] When the comprehensive maintenance value is in the third maintenance range, the third maintenance command is triggered: simultaneously replace the gear oil and gears;

[0139] When the comprehensive maintenance value is in the fourth maintenance range, the fourth maintenance instruction is triggered: do not replace the gear oil and gears.

[0140] By dividing the comprehensive maintenance value into four intervals and corresponding to different maintenance instructions, precise adaptation of pontoon winch maintenance measures is achieved. When the comprehensive maintenance value is in different intervals, instructions for oil change alone, gear change alone, simultaneous replacement, or no replacement are triggered respectively. Based on the actual matching state between the hidden damage of the gears and the oil protection capability, a targeted maintenance plan can be formulated. This prevents over-maintenance from wasting resources and avoids under-maintenance from causing hidden wear accumulation, thus ensuring the effectiveness of pontoon winch maintenance.

[0141] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A marine vessel repair and maintenance system, characterized by, The method comprises the following steps: an analysis unit is configured to acquire torque signals, rotation speed signals, gear surface temperatures, and vibration frequencies of a target object in a target process in real time, analyze the preprocessed torque signals and rotation speed signals, and obtain a gear surface alteration potential value, including: obtaining time sequence characteristics of the torque signals and the rotation speed signals in the target process, dividing the target process into three stages of starting, climbing, and peak value maintaining based on the time sequence characteristics, and generating stage length proportions; based on the stage length proportions, performing stage-by-stage feature enhancement on the preprocessed torque signals and rotation speed signals, and generating stage-by-stage torque-rotation speed feature quantities; analyzing the correlation between the torque and the rotation speed in the target process according to the stage-by-stage torque-rotation speed feature quantities, and generating a torque-rotation speed linkage coefficient; obtaining structural attributes of the target object, combining the torque-rotation speed linkage coefficient and the structural attributes, and obtaining a gear surface stress fluctuation value; calculating the preprocessed torque signals and rotation speed signals, and obtaining change rates of the torque and the rotation speed; fusing the gear surface stress fluctuation value and the change rates, and obtaining the gear surface alteration potential value; the target object is a gear disc of a pontoon winch, and the target process is a process in which the winch gear disc switches from zero speed to peak torque; a calculation unit is configured to calculate the gear surface alteration potential value according to the gear surface temperatures and the vibration frequencies, and obtain an alteration conductivity; a processing unit is configured to acquire image data of gear oil in the target object in real time, analyze the preprocessed image data, and generate an energy decay coefficient; an adaptation unit is configured to match the alteration conductivity and the energy decay coefficient, and obtain a maintenance imbalance index; a maintenance unit is configured to acquire usage lengths of the gears and the gear oil in the target object in real time, calculate the maintenance imbalance index according to the usage lengths of the gears and the gear oil, and obtain a comprehensive maintenance value; an execution unit is configured to determine a maintenance instruction according to the comprehensive maintenance value.

2. A marine maintenance system according to claim 1, wherein, The torque-rotation speed linkage coefficient and the structural attributes are combined to calculate the gear surface stress fluctuation value, including: analyzing the structural attributes, assigning weights according to force sensitivity degrees of the target object in the target process, and generating structural dynamic feature values; analyzing variation laws of the structural dynamic feature values and the torque-rotation speed linkage coefficient in each stage, and obtaining a structure-linkage dynamic coupling coefficient; based on the structural attributes, calculating deformation amounts of different meshing positions of the gear when in contact, and obtaining gear contact deformation amounts; calibrating the structure-linkage dynamic coupling coefficient according to the gear contact deformation amounts, and generating the gear surface stress fluctuation value.

3. A marine maintenance system according to claim 1, wherein, The gear surface alteration potential value is calculated according to the gear surface temperatures and the vibration frequencies, and an alteration conductivity is obtained, including: based on the gear surface temperatures, analyzing occurrence frequencies and duration lengths of temperature surges in the target process, and generating thermal shock feature values; based on the vibration frequencies, extracting jump amplitudes and duration periods of abnormal vibrations in the target process, and obtaining vibration frequency abnormality feature values; obtaining micro-texture structure parameters of the gear surface, calculating the thermal shock feature values and the vibration frequency abnormality feature values according to the micro-texture structure parameters, and obtaining thermal shock indexes and vibration frequency abnormality indexes; analyzing the synchronism of the thermal shock indexes and the vibration frequency abnormality indexes on a time axis, and generating a thermal vibration coordination factor.

4. A marine maintenance system as claimed in claim 3, wherein, According to the tooth surface temperature and vibration frequency, the tooth surface alteration potential value is calculated to obtain the alteration conductivity, and the alteration conductivity also includes: The mapping relationship between the thermal vibration synergy factor and the tooth surface alteration potential value is analyzed to obtain the synergy-alteration correlation degree; The influence value of different tooth surface temperatures and vibration frequencies on the gear is calculated, and the synergy-alteration correlation degree is corrected according to the influence value to obtain the alteration conductivity.

5. A marine maintenance system as claimed in claim 3, wherein, The pre-processed image data is analyzed to generate the energy attenuation coefficient, including: The image data is analyzed to capture the discrete trajectory and aggregation frequency of the internal microstructure of the oil, and the oil phase dispersion index is generated; The light and shadow gradient change of the oil film in the image data when the gear is engaged is calculated to obtain the oil film boundary stability value; The correlation between the oil phase dispersion index and the oil film boundary stability value in the gear operation cycle is analyzed to generate the efficiency attenuation coupling value; The morphological change amplitude and speed of the oil in the image data with the change of torque and rotating speed are analyzed to generate the working condition adaptation variation rate; The efficiency attenuation coupling value and the working condition adaptation variation rate are fused to obtain the energy attenuation coefficient.

6. A marine maintenance system as claimed in claim 5, wherein, The alteration conductivity and the energy attenuation coefficient are matched to obtain the maintenance imbalance index, including: The feature dimension of the alteration conductivity and the energy attenuation coefficient is respectively expanded to generate the alteration feature vector and the energy feature vector; The image data is analyzed to obtain the gear oil viscosity, and the gear oil viscosity is combined with the micro-texture structure parameter to obtain the viscosity-roughness adaptation value; The spatial matching degree of the alteration feature vector and the energy feature vector is calculated to obtain the alteration-energy matching deviation value; The decay rate of the lubrication effect with the degree of alteration of the gear tooth surface is calculated according to the alteration conductivity to generate the lubrication decay rate value; The viscosity-roughness adaptation value and the lubrication decay rate value are taken as weight parameters to weight and adjust the alteration-energy matching deviation value to generate the dynamic adaptation imbalance degree; Based on the dynamic adaptation imbalance degree, the peak value interval of the torque signal and the rotating speed signal is mapped to generate the maintenance imbalance index.

7. A marine maintenance system as claimed in claim 6, wherein, Based on the dynamic adaptation imbalance degree, the peak value interval of the torque signal and the rotating speed signal is mapped to generate the maintenance imbalance index, including: The transient peak variation characteristic value is generated by analyzing the transient mutation characteristics in the peak value interval of the torque signal and the rotating speed signal; The interaction of the transient peak variation characteristic value and the dynamic adaptation imbalance degree in each stage of the target process is analyzed to obtain the space-time coupling imbalance coefficient; Based on the image data, the oil film rupture critical value when the gear is engaged is calculated; The space-time coupling imbalance coefficient is dynamically calibrated according to the oil film rupture critical value to generate the calibrated space-time imbalance value; The space-time imbalance value is converted to generate the maintenance imbalance index.

8. A marine maintenance system according to claim 7, wherein, The maintenance imbalance index is calculated according to the service life of the gear and the gear oil to obtain the comprehensive maintenance value, including: The performance decay of the gear and the gear oil in different service life intervals is analyzed to generate the gear aging nonlinear decay coefficient and the oil aging nonlinear decay coefficient; The dynamic change rate of the gear aging nonlinear decay coefficient and the oil aging nonlinear decay coefficient is calculated to obtain the time length-decay synergy coefficient; The maintenance imbalance index is adaptively adjusted according to the time length-decay synergy coefficient to generate the dynamic maintenance estimation value; The tooth surface stress fluctuation value, the thermal shock characteristic value, the vibration frequency abnormality index and the oil film rupture critical value are fused to generate a multi-dimensional attenuation factor; The deviation of the dynamic maintenance estimated value is corrected according to the multi-dimensional attenuation factor to obtain a comprehensive maintenance value.

9. A marine maintenance system as claimed in claim 8, characterised in that, According to the comprehensive maintenance value, the maintenance instruction is determined, including: Based on the comprehensive maintenance value, the maintenance interval and the maintenance instruction are determined.

Images