Big data driven engine maintenance management method and system

Abstract

The invention discloses a big-data-driven engine maintenance management method and system, and relates to the technical field of maintenance management, and the method comprises the steps: carrying out the calculation of the wear rates of different working conditions on the wear data of parts with different working conditions, and obtaining the wear rates with different working conditions; according to the working condition difference wear rate, stress increment conduction analysis between component connection points is conducted, connection point stress increment conduction data are obtained, connection point failure critical interval recognition is conducted based on the connection point stress increment conduction data, and a connection point failure critical interval is obtained; according to the connection point failure critical interval, maintenance controllable strategy adjustment between different working conditions is carried out on the working condition difference part abrasion data, and maintenance strategy adjustment normalized data is obtained; and carrying out maintenance learning on the connection point failure critical interval and the maintenance strategy adjustment normalization data to obtain maintenance control logic data, and carrying out logic firmware design based on the maintenance control logic data to obtain maintenance control logic firmware. The engine maintenance management method has the effect of improving the engine maintenance management efficiency.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of maintenance management, in particular to an engine maintenance management method and system driven by big data. BACKGROUND

[0002] Engine maintenance management is the core support link in the fields of modern aviation, machinery, etc., which controls the whole process of inspection, maintenance, repair and overhaul of the engine through a scientific and rigorous system. Its core goal is to ensure that the engine is always in a safe and reliable airworthy or running state, and effectively prolong its service life.

[0003] In related technologies, the existing engine maintenance management mainly relies on periodic maintenance plans or responsive maintenance based on simple fault codes, lacks real-time monitoring and in-depth data analysis of the engine operating state, and this method has the following defects: First, the fixed maintenance period cannot adapt to the actual wear of the engine in complex operating environments, resulting in over-maintenance or insufficient maintenance, increasing operating costs and safety risks; Second, the existing engine maintenance management can only handle explicit faults and cannot predict potential fault points. For example, under high temperature, high load or frequent start-stop conditions, the progressive wear of internal components of the engine is often ignored until a serious failure occurs; Third, the existing technology lacks fusion analysis of multi-source data, resulting in one-sided maintenance decisions, which cannot comprehensively evaluate the engine health status, thereby reducing the efficiency of engine maintenance management, and there is room for improvement. SUMMARY

[0004] In view of the deficiencies of the prior art, the present application provides an engine maintenance management method and system driven by big data.

[0005] In a first aspect, the present application provides an engine maintenance management method driven by big data, comprising the following steps: Step S1: Obtain engine operation log data, extract working condition environment data and engine component state data from the engine operation log data, respectively obtain working condition environment data and engine component state data, map the component dynamic wear between different working conditions according to the working condition environment data, and obtain working condition difference component wear data; Step S2: Calculate the wear rate between different working conditions for the working condition difference component wear data, obtain the working condition difference wear rate, conduct stress increment conduction analysis between component connection points according to the working condition difference wear rate, obtain connection point stress increment conduction data, identify the connection point failure critical interval based on the connection point stress increment conduction data, and obtain the connection point failure critical interval; Step S3: Based on the critical failure range of the connection point, adjust the controllable maintenance strategy for different working conditions of the wear data of components with different working conditions to obtain normalized data of maintenance strategy adjustment; Step S4: Perform maintenance control logic learning on the critical interval of connection point failure and the normalized data of maintenance strategy adjustment to obtain maintenance control logic data. Based on the maintenance control logic data, design automated logic firmware to obtain maintenance control logic firmware. Embed the maintenance control logic firmware into the engine monitoring terminal to perform engine maintenance management.

[0006] Preferably, step S1 includes the following steps: Step S11: Obtain engine operation log data, and perform data cleaning and outlier processing on the engine operation log data to obtain cleaned operation log data; Step S12: Extract operating environment data and engine component status data from the cleaned operation log data respectively to obtain operating environment data and engine component status data. Step S13: Perform time-series alignment and interpolation on the engine component status data to obtain missing component status data; Step S14: Based on the working environment data, perform dynamic wear mapping of components under different working conditions to fill in missing component status data and obtain wear data of components with different working conditions.

[0007] Preferably, step S2 includes the following steps: Step S21: Calculate the instantaneous wear rate between different working conditions for the wear data of components with different working conditions, and obtain the wear rate with different working conditions; Step S22: Simulate the instantaneous thermal stress response of the component based on the wear rate difference under working conditions to obtain the instantaneous thermal stress response data of the component; Step S23: Obtain engine design data, which includes component material properties, connection structure geometric parameters, and thermal conductivity coefficient; Step S24: Based on the wear rate under different operating conditions and the instantaneous thermal stress response data of the components, perform stress increment transmission analysis between component connection points on the engine design data to obtain stress increment transmission data at the connection points; Step S25: Identify stress superposition connection points by analyzing the stress increment transmission data at the connection points to obtain stress superposition connection points; Step S26: Based on the stress increment transmission data of the connection point and the engine design data, identify the critical failure interval of the connection point for stress superposition and obtain the critical failure interval of the connection point.

[0008] Preferably, step S24 includes the following steps: Step S241: Perform topology connection structure analysis on the engine design data to obtain engine topology connection structure data; Step S242: Perform thermal stress flow guidance analysis on the engine topology connection structure data based on the instantaneous thermal stress response data of the components to obtain the thermal stress flow guidance data of the connection structure; Step S243: Analyze the temperature difference distribution at the connection point based on the thermal stress flow guidance data of the connection structure to obtain the temperature difference distribution data at the connection point; Step S244: Calculate the local connection point temperature difference increment series based on the wear rate of the working condition difference to obtain the local connection point temperature difference increment series; Step S245: Based on the local connection point temperature difference increment series and connection point temperature difference distribution data, perform stress increment transmission analysis between component connection points on the engine topology connection structure data to obtain connection point stress increment transmission data.

[0009] Preferably, step S26 includes the following steps: Step S261: Analyze the material properties of the component connection lines in the engine design data to obtain the material property data of the component connection lines; Step S262: Calculate the stress vector of the stress superimposed connection point based on the stress increment transmission data of the connection point to obtain the stress vector data of the superimposed connection point; Step S263: Based on the stress vector data of the superimposed connection point, perform thermal expansion plastic strain fitting on the material property data of the component connection line to obtain thermal expansion plastic strain fitting data; Step S264: Calculate the fatigue tolerance of the superimposed connection point by performing stress vector data on the superimposed connection point to obtain fatigue tolerance data of the superimposed connection point; Step S265: Based on the thermal expansion plastic strain fitting data and the fatigue tolerance data of the superimposed points, identify the critical interval of connection point failure for the stress superimposed connection points, and obtain the critical interval of connection point failure.

[0010] Preferably, step S263 includes the following steps: Step S2631: Extract the heat capacity of the component connection circuit material from the material property data to obtain the heat capacity of the circuit material; Step S2632: Calculate the average difference of thermal energy in the stress direction on the stress vector data of the superimposed connection point to obtain the average difference of thermal energy in the stress direction; Step S2633: Simulate the plastic yield limit of the heat capacity of the line material based on the average difference of thermal energy in the stress direction, and obtain the plastic yield data of the material with average difference of thermal energy. Step S2634: Based on the superimposed connection point stress vector data, perform plastic anisotropy analysis on the thermal energy uniformity material plastic yield data to obtain the material plastic yield anisotropy data; Step S2635: Based on the thermal energy uniformity difference material plastic yield data, the material plastic yield anisotropy data, and the stress direction thermal energy uniformity difference data, perform thermal expansion plastic strain fitting to obtain thermal expansion plastic strain fitting data.

[0011] Preferably, step S264 includes the following steps: Step S2641: Analyze the change in heat flux density at the connection point by performing stress vector data at the superimposed connection point to obtain the change data of heat flux density at the connection point; Step S2642: Perform density increment trend analysis on the heat flux density change data at the connection point to obtain heat flux density increment trend data; Step S2643: Based on the heat flux density increment trend data, perform piecewise mapping of the fatigue gradient fluctuation data at the connection point to obtain the fatigue gradient piecewise fluctuation data. Step S2644: Perform fatigue limit numerical calculation on the fatigue gradient segmented fluctuation data to obtain the fatigue limit value; Step S2645: Based on the fatigue limit value and the fatigue gradient piecewise fluctuation data, calculate the fatigue tolerance of the superimposed connection point stress vector data to obtain the fatigue tolerance data of the superimposed point.

[0012] Preferably, step S3 includes the following steps: Step S31: Analyze the lower limit safe operating threshold between different operating conditions based on the wear data of components with different operating conditions, and obtain the lower limit safe operating threshold between different operating conditions; Step S32: Based on the critical interval of connection point failure, adjust the lower limit safe operation threshold between different road conditions to obtain maintenance controllable strategy adjustment data. Step S33: Normalize the maintenance controllable strategy adjustment data to obtain normalized maintenance strategy adjustment data.

[0013] Preferably, step S32 includes the following steps: Step S321: Based on the critical interval of connection point failure, calculate the upper limit of instantaneous maintenance capacity under different working conditions for the lower limit safe operation threshold of different road conditions, and obtain the upper limit data of instantaneous maintenance capacity; Step S322: Based on the critical interval of connection point failure, analyze the controllable maintenance limit time under different working conditions of the instantaneous maintenance load limit data to obtain the controllable maintenance limit time data; Step S323: Perform maintenance resource load balancing processing based on instantaneous maintenance capacity limit data and maintenance capacity controllable duration data to obtain maintenance resource load balancing data; Step S324: Based on the instantaneous maintenance capacity limit data, the maintenance limit continuous controllable time data, and the maintenance resource load balance data, adjust the maintenance controllable strategy for different working conditions according to the lower limit safe operation threshold for different road conditions, and obtain maintenance controllable strategy adjustment data.

[0014] Secondly, this application provides a big data-driven engine maintenance management system, including: The data acquisition module is used to acquire engine operation log data, extract operating environment data and engine component status data from the engine operation log data, and obtain operating environment data and engine component status data respectively. Based on the operating environment data, the module performs dynamic wear mapping of components between different operating conditions on the engine component status data to obtain wear data of components with different operating conditions. The analysis module is used to calculate the wear rate between different working conditions based on the wear data of components with different working conditions, obtain the wear rate of different working conditions, perform stress increment transmission analysis between component connection points based on the wear rate of different working conditions, obtain stress increment transmission data of connection points, and identify the critical failure interval of connection points based on the stress increment transmission data of connection points, thereby obtaining the critical failure interval of connection points. The adjustment module is used to adjust the controllable maintenance strategy between different working conditions based on the wear data of components with different working conditions according to the critical range of connection point failure, and to obtain normalized data for maintenance strategy adjustment. The management module is used to learn the maintenance control logic from the critical interval of connection point failure and the normalized data of maintenance strategy adjustment to obtain maintenance control logic data. Based on the maintenance control logic data, the module designs automated logic firmware to obtain maintenance control logic firmware. The maintenance control logic firmware is then embedded into the engine monitoring terminal to perform engine maintenance management.

[0015] In summary, this application includes the following beneficial technical effects: This application provides a big data-driven engine maintenance management method. By deeply analyzing and mapping engine operation log data, it can identify potential engine fault points in real time, provide early warnings of wear or failure risks, and reduce sudden failures and downtime. This effectively reduces the occurrence of over-maintenance or under-maintenance caused by fixed maintenance cycles failing to adapt to the actual wear and tear of the engine under complex operating environments. Secondly, based on data-driven maintenance strategy adjustments, maintenance plans and resource allocation are optimized, reducing maintenance costs and human error. Thirdly, through automated logic learning firmware, real-time response and adaptive control of maintenance management are achieved, improving engine reliability and service life, thereby effectively improving engine maintenance management efficiency. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a flowchart of a big data-driven engine maintenance management method according to an embodiment of this application.

[0018] Figure 2 This is a schematic diagram of a big data-driven engine maintenance management system according to an embodiment of this application. Detailed Implementation

[0019] The following is in conjunction with the appendix Figures 1-2 This application will be described in further detail.

[0020] Example 1 This application discloses a big data-driven engine maintenance management method.

[0021] Reference Figure 1 A big data-driven engine maintenance management method includes the following steps: Step S1: Obtain engine operation log data, extract operating environment data and engine component status data from the engine operation log data to obtain operating environment data and engine component status data respectively, and perform dynamic wear mapping of components between different operating conditions based on the operating environment data to obtain wear data of components with different operating conditions. Step S2: Calculate the wear rate between different working conditions for the wear data of components with different working conditions to obtain the wear rate of different working conditions. Perform stress increment transmission analysis between component connection points based on the wear rate of different working conditions to obtain the stress increment transmission data of connection points. Identify the critical failure interval of connection points based on the stress increment transmission data of connection points to obtain the critical failure interval of connection points. Step S3: Based on the critical failure range of the connection point, adjust the controllable maintenance strategy for different working conditions of the wear data of components with different working conditions to obtain normalized data of maintenance strategy adjustment; Step S4: Perform maintenance control logic learning on the critical interval of connection point failure and the normalized data of maintenance strategy adjustment to obtain maintenance control logic data. Based on the maintenance control logic data, design automated logic firmware to obtain maintenance control logic firmware. Embed the maintenance control logic firmware into the engine monitoring terminal to perform engine maintenance management.

[0022] Specifically, in step S4, a maintenance control logic learning framework is first constructed. The critical interval of connection point failure is concatenated with the maintenance strategy adjustment data, and a deep Q-learning algorithm is used to train the maintenance decision model. The optimal control strategy is obtained by iteratively optimizing the reward function. Subsequently, the automated logic firmware is designed. The learned control strategy is compiled into embedded executable code. A modular architecture is used to design three core modules: state monitoring, decision reasoning, and control output. The reliability of the logic firmware under real-time constraints is verified through hardware-in-the-loop testing. Finally, the maintenance control logic firmware is burned into the FPGA chip of the engine monitoring terminal, and a lightweight inference engine is deployed in the embedded system to realize closed-loop automated management from data acquisition to maintenance decision-making.

[0023] It should be noted that step S1 includes the following steps: Step S11: Obtain engine operation log data, and perform data cleaning and outlier processing on the engine operation log data to obtain cleaned operation log data; Step S12: Extract operating environment data and engine component status data from the cleaned operation log data respectively to obtain operating environment data and engine component status data. Step S13: Perform time-series alignment and interpolation on the engine component status data to obtain missing component status data; Step S14: Based on the working environment data, perform dynamic wear mapping of components under different working conditions to fill in missing component status data and obtain wear data of components with different working conditions.

[0024] Specifically, in step S11, engine operation log data is first acquired through the data acquisition interface of the engine monitoring unit. A sliding window-based outlier detection algorithm is used to clean the data. Specifically, abnormal fluctuations are identified by calculating the ratio of the standard deviation to the median of each parameter within the time window. For identified outliers, the weighted average of adjacent data points is used for replacement. Simultaneously, a time series stationarity test method is used to eliminate periodic noise during data acquisition, ultimately obtaining cleaned operation log data containing complete timestamps. In step S12, the cleaned operation log data undergoes multi-dimensional feature separation processing. For operating environment data extraction, a density-based clustering algorithm is used to segment operating conditions. By calculating the cluster centers of environmental parameters in the feature space, typical operating condition modes, including high temperature and high humidity, and low temperature and low pressure, are identified, and environmental feature vectors within each operating condition segment are extracted. For engine component status data extraction, multi-source sensor data fusion technology is used. After time alignment of the monitoring data from vibration sensors, temperature sensors, and pressure sensors, the statistical characteristics of each sensor reading within the sliding window are calculated. In step S13, the extracted component state data undergoes time series integrity processing. First, a time series database indexed by the acquisition time is established. For data points with inconsistent acquisition intervals, a cubic spline interpolation algorithm is used for time axis normalization. For missing data segments, missing values ​​are predicted based on an autoregressive moving average model. The rationality of the interpolation results is verified by calculating the Euclidean distance between the predicted values ​​and adjacent measured values, ensuring that all state parameters have a uniform time resolution and data length. In step S14, dynamic wear mapping analysis between operating conditions is performed based on the complete state data. First, a correlation matrix between operating environment parameters and component state parameters is constructed. By calculating the gradient of the change in component vibration spectrum characteristics during different operating condition transitions, the correspondence between operating condition switching and wear accumulation is established. An operating condition identification method based on a hidden Markov model is used, combined with the fatigue characteristic curve of the component material, to calculate the equivalent wear amount in each operating condition stage. Finally, differential component wear data containing operating condition identifiers is generated. This data accurately reflects the wear distribution characteristics of various engine components under different environmental conditions.

[0025] It should be noted that step S2 includes the following steps: Step S21: Calculate the instantaneous wear rate between different working conditions for the wear data of components with different working conditions, and obtain the wear rate with different working conditions; Step S22: Simulate the instantaneous thermal stress response of the component based on the wear rate difference under working conditions to obtain the instantaneous thermal stress response data of the component; Step S23: Obtain engine design data, which includes component material properties, connection structure geometric parameters, and thermal conductivity coefficient; Step S24: Based on the wear rate under different operating conditions and the instantaneous thermal stress response data of the components, perform stress increment transmission analysis between component connection points on the engine design data to obtain stress increment transmission data at the connection points; Step S25: Identify stress superposition connection points by analyzing the stress increment transmission data at the connection points to obtain stress superposition connection points; Step S26: Based on the stress increment transmission data of the connection point and the engine design data, identify the critical failure interval of the connection point for stress superposition and obtain the critical failure interval of the connection point.

[0026] Specifically, in step S21, time series analysis is first performed on the wear data of components with different operating conditions. The instantaneous wear rate is obtained by calculating the rate of change of wear depth at adjacent time points. Specifically, the sliding window differential method is used to calculate the first derivative of wear amount with respect to time in each operating condition segment. Combined with the identification of operating condition switching points, the rate curve is smoothed to eliminate noise interference, thereby obtaining the operating condition-differential wear rate that reflects the dynamic changes of wear under different operating conditions. In step S22, the instantaneous thermal stress response of the component is simulated based on the calculated operating condition-differential wear rate. The thermoelastic coupling analysis method is used to calculate the effect of wear-induced surface deformation on heat distribution by establishing a correlation model between the component's temperature field and stress field. The impact is analyzed by using finite element method (FEM) software to simulate the thermal expansion effect of components under transient temperature changes, and combining this with the thermophysical parameters of the materials to output instantaneous thermal stress distribution data of the components under different wear rates. In step S23, key design parameters are obtained from the engine design database, including component material properties (such as elastic modulus, yield strength, and coefficient of thermal expansion), connection structure geometric parameters (such as bolt preload, contact surface dimensions, and fit tolerances), and thermal conductivity (describing the efficiency of heat transfer between materials). These data serve as the baseline input for subsequent stress analysis. In step S24, by combining the wear rate under different operating conditions and the instantaneous thermal stress response data of the components, the stress increment transmission between component connection points is analyzed in the engine design data. First, the topology of the connected network is constructed. By calculating the transfer function of stress waves between the connection nodes, the propagation path of stress increments caused by wear and thermal stress superposition is analyzed. The dynamic load spectrum method is used to simulate the attenuation and superposition effects of stress between multiple connection points. Based on the material constitutive relation, the stress concentration factor of each connection point is calculated, and finally, the stress increment transmission data of the connection points is generated. In step S25, stress superposition connection points are identified on the stress increment transmission data of the connection points. By calculating the principal stress vector sum at each connection point and combining it with historical stress data, an anomaly detection algorithm based on cluster analysis is used to identify superposition regions where the stress value exceeds a certain proportion (such as a multiple of the standard deviation) of the average stress level. At the same time, the topology of the connected network is constructed. By calculating the transfer function of stress waves between the connection nodes, the propagation path of stress increments caused by wear and thermal stress superposition is analyzed. The equivalent stress under multiaxial stress conditions is calculated to identify key stress superposition connection points. In step S26, based on the stress increment transmission data of the connection points and engine design data, the critical failure interval of the stress superposition connection points is identified. Using reliability engineering methods, the correlation curve between material fatigue life and stress amplitude is calculated. Combined with the number of stress cycles and the safety factor, the failure probability distribution of the connection points is determined. The historical failure data is trained using a support vector machine classification model. Based on the stress concentration factor, load frequency, and material toughness, the safe operation interval, risk warning interval, and critical failure interval are divided. Finally, the critical failure interval data of the connection points is output, providing a basis for the formulation of subsequent maintenance strategies.

[0027] It should be noted that step S24 includes the following steps: Step S241: Perform topology connection structure analysis on the engine design data to obtain engine topology connection structure data; Step S242: Perform thermal stress flow guidance analysis on the engine topology connection structure data based on the instantaneous thermal stress response data of the components to obtain the thermal stress flow guidance data of the connection structure; Step S243: Analyze the temperature difference distribution at the connection point based on the thermal stress flow guidance data of the connection structure to obtain the temperature difference distribution data at the connection point; Step S244: Calculate the local connection point temperature difference increment series based on the wear rate of the working condition difference to obtain the local connection point temperature difference increment series; Step S245: Based on the local connection point temperature difference increment series and connection point temperature difference distribution data, perform stress increment transmission analysis between component connection points on the engine topology connection structure data to obtain connection point stress increment transmission data.

[0028] Specifically, in step S241, a topological connection structure analysis is first performed on the engine design data. This involves constructing a connection relationship graph of each engine component, identifying key connection nodes using a node degree centrality algorithm from graph theory, and calculating the stiffness coefficients and constraints of each connection point based on the contact surface geometric parameters. Simultaneously, an adjacency matrix of the connection structure is established by analyzing the assembly relationships between components, ultimately generating engine topological connection structure data containing connection types, load transfer paths, and boundary conditions. In step S242, a thermal stress flow guidance analysis is performed on the engine topological connection structure data based on the instantaneous thermal stress response data of the components, using thermoelasticity principles... A coupled model of temperature and stress fields is established. The transmission efficiency of thermal stress in the connection network is calculated, and stress concentration is analyzed based on the difference in the thermal expansion coefficients of the materials. Vector field analysis is used to determine the main direction of thermal stress transmission, ultimately outputting thermal stress flow guidance data describing the distribution of thermal stress in the connection structure. In step S243, the thermal stress flow guidance data of the connection structure is analyzed for temperature difference distribution at connection points. By establishing the heat balance equation of the connection points, the temperature gradient between adjacent connection points is calculated. A node temperature prediction method based on thermal network theory is used, combined with the thermal conductivity characteristics of the materials and boundary heat exchange conditions, to analyze the transient operating conditions. The temperature change rate at each connection point is calculated, and a continuous temperature distribution cloud map is obtained by spatiotemporal interpolation of the temperature field, ultimately generating connection point temperature difference distribution data reflecting the temperature difference characteristics between connection points. In step S244, the local connection point temperature difference increment series is calculated based on the wear rate of the working condition difference. First, a correlation model between the wear rate and the change in thermal resistance is established. By calculating the degree of decrease in heat conduction efficiency of the contact surface caused by wear, the cumulative effect of local temperature difference is analyzed. Then, a series expansion method is used to perform polynomial fitting on the temperature difference increment. The development trend of temperature difference is predicted by combining the time function of the wear process, and the temperature difference between adjacent time steps is calculated. The rate of change determines the increment of temperature difference at local connection points. In step S245, stress increment transmission between component connection points is analyzed based on the increment of temperature difference at local connection points and the temperature difference distribution data at connection points. By constructing a propagation model of stress waves in the connection network, the transmission process of thermal stress increment caused by temperature difference between nodes is calculated. The interaction of multi-source stress at connection points is analyzed using the dynamic load superposition principle, and the stress relaxation effect is evaluated in combination with the material creep characteristics. Finally, the accuracy of the stress transmission path is verified by finite element simulation, and the stress increment transmission data of connection points containing stress concentration coefficient and fatigue damage parameters is output.

[0029] It should be noted that step S26 includes the following steps: Step S261: Analyze the material properties of the component connection lines in the engine design data to obtain the material property data of the component connection lines; Step S262: Calculate the stress vector of the stress superimposed connection point based on the stress increment transmission data of the connection point to obtain the stress vector data of the superimposed connection point; Step S263: Based on the stress vector data of the superimposed connection point, perform thermal expansion plastic strain fitting on the material property data of the component connection line to obtain thermal expansion plastic strain fitting data; Step S264: Calculate the fatigue tolerance of the superimposed connection point by performing stress vector data on the superimposed connection point to obtain fatigue tolerance data of the superimposed connection point; Step S265: Based on the thermal expansion plastic strain fitting data and the fatigue tolerance data of the superimposed points, identify the critical interval of connection point failure for the stress superimposed connection points, and obtain the critical interval of connection point failure.

[0030] Specifically, in step S261, the material properties of the component connection lines in the engine design data are first analyzed. Constitutive parameters of each component material in the connection structure, including elastic modulus, Poisson's ratio, yield strength, and coefficient of thermal expansion, are extracted. The equivalent mechanical properties of anisotropic materials are calculated using material microstructure analysis methods, and the microstructure stability is evaluated based on material heat treatment process records. Simultaneously, the hardening curve of the material under cyclic loading is obtained by fitting experimental data, ultimately generating material property data for the component connection lines that includes stress-strain response characteristics and fatigue life parameters. In step S262, stress stacking is analyzed based on the stress increment transmission data at the connection points. Stress vector calculations are performed at the connection points. Tensor analysis is used to decompose the multiaxial stress state at each connection point into principal stress components. The direction of the combined stress is determined by calculating the magnitude and direction angle of the stress vector. The phase relationship of peak stress occurrence is identified by combining stress time history data. Finally, superimposed connection point stress vector data describing the magnitude, direction, and distribution characteristics of stress at the superimposed connection points are output. In step S263, thermal expansion plastic strain fitting is performed on the material property data of the component connection lines based on the superimposed connection point stress vector data to establish an elasto-plastic constitutive model considering temperature effects. The coupling effect of thermal stress and mechanical stress is calculated to analyze the material... The plastic flow law of the material under complex stress state is studied, and the strain increment caused by thermal expansion is solved by iterative algorithm. Finally, thermal expansion plastic strain fitting data reflecting the deformation behavior of the material under thermomechanical load is obtained. In step S264, the fatigue tolerance of the superimposed connection point is calculated on the stress vector data of the superimposed connection point. Based on the joint analysis method of stress life curve and strain life curve, the fatigue life under multiaxial stress state is evaluated by calculating the combined effect of equivalent alternating stress amplitude and average stress, combined with the material SN curve and Miner's cumulative damage rule. The influence of surface processing quality and service environment factors is considered and correction coefficients are introduced. The final output includes safety The fatigue tolerance data at the superposition point of full margin and life prediction values; in step S265, the critical interval of connection point failure is identified based on the thermal expansion plastic strain fitting data and the fatigue tolerance data at the superposition point, the failure assessment map is established using the limit state design principle, the critical damage threshold is determined by calculating the interaction between plastic strain energy and fatigue damage, the failure probability distribution is analyzed using reliability theory, and the accuracy of the critical interval is verified by combining historical failure case data. Finally, the critical interval data of connection point failure, which includes the safe operation interval, the risk warning interval, and the critical failure interval, is output, providing a quantitative basis for the assessment of engine structural integrity.

[0031] It should be noted that step S263 includes the following steps: Step S2631: Extract the heat capacity of the component connection circuit material from the material property data to obtain the heat capacity of the circuit material; Step S2632: Calculate the average difference of thermal energy in the stress direction on the stress vector data of the superimposed connection point to obtain the average difference of thermal energy in the stress direction; Step S2633: Simulate the plastic yield limit of the heat capacity of the line material based on the average difference of thermal energy in the stress direction, and obtain the plastic yield data of the material with average difference of thermal energy. Step S2634: Based on the superimposed connection point stress vector data, perform plastic anisotropy analysis on the thermal energy uniformity material plastic yield data to obtain the material plastic yield anisotropy data; Step S2635: Based on the thermal energy uniformity difference material plastic yield data, the material plastic yield anisotropy data, and the stress direction thermal energy uniformity difference data, perform thermal expansion plastic strain fitting to obtain thermal expansion plastic strain fitting data.

[0032] Specifically, in step S2631, the material heat capacity is first extracted from the component connection line material property data. By analyzing the specific heat capacity change curves of the material at different temperatures, and combining the atomic heat capacity weighted calculation method of the material's constituent elements, dynamic heat capacity parameters that change with temperature are obtained. The influence of latent heat during the material phase transition process is also considered, and finally, the heat capacity data of the line material that accurately reflects the material's heat storage capacity is obtained. In step S2632, the stress direction thermal energy difference is calculated from the stress vector data of the superimposed connection points. The thermoelastic energy analysis method is used to calculate the heat-work conversion in the principal stress direction. Efficiency differences are analyzed by combining the distribution characteristics of stress vector magnitude and direction, and the local heat energy accumulation phenomenon caused by different stress paths is analyzed. Based on the first law of thermodynamics, the non-uniformity of heat energy distribution in each stress direction is calculated, and finally, stress direction heat energy average difference data describing the coupling characteristics of the stress field and temperature field is output. In step S2633, the plastic yield limit of the circuit material's heat capacity is simulated based on the stress direction heat energy average difference data. A modified yield criterion model considering the temperature softening effect is established. By calculating the degree of local softening of the material caused by the non-uniform heat energy distribution, and combining the dynamic recrystallization critical condition analysis, the simulation is performed. The evolution of the yield behavior of the material under complex thermomechanical loads ultimately yields thermal energy uniform difference material plastic yield data reflecting the variation law of the material's yield strength under the influence of the temperature field. In step S2634, plastic anisotropy analysis is performed on the thermal energy uniform difference material plastic yield data based on the superimposed connection point stress vector data. Using the framework of crystal plasticity theory, the direction dependence of the material's plastic deformation is analyzed by calculating the number of slip systems activated in different stress directions. Combined with the texture evolution model, the anisotropic strengthening effect induced by plastic strain is predicted, and finally, the difference in plastic response of the material under different loading directions is output. The material plastic yield anisotropy data are obtained; in step S2635, thermal expansion plastic strain is fitted based on the thermal energy uniformity difference material plastic yield data, material plastic yield anisotropy data, and stress direction thermal energy uniformity difference data to establish a unified constitutive model framework. The coupling relationship between thermal expansion strain and plastic strain is solved iteratively. The plastic strain increment under non-proportional loading path is calculated by combining the anisotropic yield criterion. The model accuracy is verified by fitting experimental data using a nonlinear regression method. Finally, thermal expansion plastic strain fitting data that can accurately predict the deformation behavior of materials under complex thermomechanical loads are output.

[0033] It should be noted that step S264 includes the following steps: Step S2641: Analyze the change in heat flux density at the connection point by performing stress vector data at the superimposed connection point to obtain the change data of heat flux density at the connection point; Step S2642: Perform density increment trend analysis on the heat flux density change data at the connection point to obtain heat flux density increment trend data; Step S2643: Based on the heat flux density increment trend data, perform piecewise mapping of the fatigue gradient fluctuation data at the connection point to obtain the fatigue gradient piecewise fluctuation data. Step S2644: Perform fatigue limit numerical calculation on the fatigue gradient segmented fluctuation data to obtain the fatigue limit value; Step S2645: Based on the fatigue limit value and the fatigue gradient piecewise fluctuation data, calculate the fatigue tolerance of the superimposed connection point stress vector data to obtain the fatigue tolerance data of the superimposed point.

[0034] Specifically, in step S2641, the heat flux density variation at the superimposed connection point stress vector data is first analyzed. A two-way coupled model of the stress and temperature fields is established, and Fourier's law of heat conduction is used to calculate the heat flux distribution characteristics in the stress concentration region. Combined with the material's thermal diffusivity and specific heat capacity parameters, the dynamic change process of heat flux density under cyclic loading is analyzed. Based on the angle between the stress vector direction and the heat flux direction, the spatial distribution gradient of heat flux density is determined, ultimately obtaining the connection point heat flux density variation data reflecting the change in heat flux at the connection point over time. In step S2642, the density increment trend of the connection point heat flux density variation data is analyzed. Potential analysis employs a time series prediction method. By calculating the rate of change of heat flux density within adjacent time windows, the stage characteristics of heat flux development are identified. Wavelet transform technology is used to extract the long-term trend component and short-term fluctuation component in the heat flux density change. A moving average algorithm is used to smooth random fluctuations to highlight the essential change law. Finally, the incremental trend data of heat flux density describing the development trend and change amplitude are output. In step S2643, the fatigue gradient fluctuation of the heat flux density change data at the connection point is segmented and mapped according to the incremental trend data of heat flux density. By identifying the inflection points and extreme points of the heat flux density change curve, the continuous heat flux density change process is divided into segments. The system is divided into segments with different characteristics. A dynamic time warping algorithm is used to analyze the correspondence between heat flux density fluctuations and fatigue damage accumulation within each segment, and a mapping model between heat flux gradient and fatigue damage gradient is established. Finally, segmented fatigue gradient fluctuation data reflecting the fatigue damage characteristics of different heat flux stages is output. In step S2644, fatigue limit numerical calculations are performed on the segmented fatigue gradient fluctuation data. Based on the material's SN curve and Miner's linear cumulative damage theory, the equivalent stress amplitude and cycle number within each fluctuation segment are calculated, and corrections are made using stress concentration factor and surface quality factor. A nonlinear fitting method is used to determine the fatigue limit of the material under specific heat flux conditions. The fatigue limit value is further corrected by considering the influence of mean stress and multiaxial stress state, and finally the accurate fatigue limit value is obtained. In step S2645, the fatigue tolerance of the superimposed connection point stress vector data is calculated based on the fatigue limit value and fatigue gradient segmented fluctuation data. A fatigue life prediction model considering the influence of heat flow is established. The safety factor is determined by calculating the ratio of the actual stress level to the fatigue limit. The non-uniformity of damage accumulation is evaluated by combining the fatigue gradient fluctuation characteristics. The fatigue tolerance range at a certain confidence level is calculated by using reliability theory. Finally, the superimposed point fatigue tolerance data containing the minimum fatigue life prediction value and the safe operation boundary is output.

[0035] It should be noted that step S3 includes the following steps: Step S31: Analyze the lower limit safe operating threshold between different operating conditions based on the wear data of components with different operating conditions, and obtain the lower limit safe operating threshold between different operating conditions; Step S32: Based on the critical interval of connection point failure, adjust the lower limit safe operation threshold between different road conditions to obtain maintenance controllable strategy adjustment data. Step S33: Normalize the maintenance controllable strategy adjustment data to obtain normalized maintenance strategy adjustment data.

[0036] Specifically, in step S31, the wear data of components with different operating conditions are first analyzed for the lower limit safe operating threshold between different operating conditions. By establishing a wear limit prediction model based on reliability theory, the statistical characteristics of wear data under each operating condition are fitted using the Weibull distribution function. The probability distribution of wear amount at different confidence levels is calculated. Combined with the correspondence between wear critical values ​​and actual failure cases in historical fault data, the minimum allowable wear amount to ensure normal system operation under each operating condition is determined by finding the intersection of the wear growth curve and the preset safety boundary. Finally, the lower limit safe operating threshold data including steady-state, transient, and extreme operating conditions are output. In step S32, the lower limit safe operating threshold between different operating conditions is adjusted according to the connection point failure critical interval. A maintenance decision matrix based on risk priority is established. The relative distance between the current wear state and the failure critical interval is calculated. The risk level is assessed, and considering maintenance resource constraints and production plan requirements, a multi-objective optimization method is used to balance the cost-effectiveness of preventive maintenance and predictive maintenance. Differentiated maintenance response strategies are developed for different risk levels, including control measures such as advancing the maintenance window, increasing inspection frequency, and adjusting maintenance depth. Finally, controllable maintenance strategy adjustment data containing time scheduling, resource allocation, and technical solutions is generated. In step S33, the controllable maintenance strategy adjustment data is normalized. The min-max standardization method is used to transform maintenance parameters of different dimensions into a unified numerical range. The weight coefficients are determined by calculating the utility function values ​​of each maintenance strategy. The importance of various maintenance activities is quantitatively evaluated by combining expert scoring and analytic hierarchy process, and a ranking rule for strategy priority is established. Finally, consistent and comparable normalized maintenance strategy adjustment data is output, providing standardized input for the subsequent development of maintenance control logic.

[0037] It should be noted that step S32 includes the following steps: Step S321: Based on the critical interval of connection point failure, calculate the upper limit of instantaneous maintenance capacity under different working conditions for the lower limit safe operation threshold of different road conditions, and obtain the upper limit data of instantaneous maintenance capacity; Step S322: Based on the critical interval of connection point failure, analyze the controllable maintenance limit time under different working conditions of the instantaneous maintenance load limit data to obtain the controllable maintenance limit time data; Step S323: Perform maintenance resource load balancing processing based on instantaneous maintenance capacity limit data and maintenance capacity controllable duration data to obtain maintenance resource load balancing data; Step S324: Based on the instantaneous maintenance capacity limit data, the maintenance limit continuous controllable time data, and the maintenance resource load balance data, adjust the maintenance controllable strategy for different working conditions according to the lower limit safe operation threshold for different road conditions, and obtain maintenance controllable strategy adjustment data.

[0038] Specifically, in step S321, the instantaneous maintenance capacity limit is first calculated based on the lower limit safe operating threshold for different operating conditions according to the critical interval of connection point failure. A maintenance capability model based on real-time risk assessment is established, and dynamic programming is used to analyze the remaining safety margin of each connection point from the critical failure point under the current wear state. Combined with the technical capabilities and equipment conditions of the maintenance team, the maximum wear that a single maintenance operation can handle is calculated under the premise of ensuring system safety. Simultaneously, the constraints of different operating conditions on maintenance operations are considered, and finally, the instantaneous maintenance capacity limit data reflecting the immediate maintenance capability limit under each operating condition is output. In step S322, the maintenance capacity limit data for the instantaneous maintenance capacity limit is analyzed based on the critical interval of connection point failure. A maintenance efficiency decay model is constructed to analyze the dynamic change law of maintenance capability under continuous operation conditions. A time series prediction method is used to calculate the time window for maintaining optimal maintenance efficiency, and the feasibility of continuous operation is evaluated in conjunction with equipment reliability data and personnel fatigue coefficients. Finally, the maximum sustainable maintenance capacity that ensures no decline in maintenance quality is obtained. The maintenance upper limit and controllable time data for operation time are obtained. In step S323, maintenance resource load balancing is performed based on instantaneous maintenance capacity upper limit data and maintenance upper limit controllable time data. A multi-objective optimization model is established. By calculating the resource demand and supply matching degree of each maintenance node, a heuristic algorithm is used to solve the optimal resource allocation scheme. At the same time, the emergency response requirements of sudden failures are considered to retain appropriate resource redundancy. The task allocation strategy is dynamically adjusted based on the load balancing principle, and finally, maintenance resource load balancing data that maximizes resource utilization is output. In step S324, maintenance controllable strategies are adjusted for the lower limit safe operation thresholds between different working conditions based on instantaneous maintenance capacity upper limit data, maintenance upper limit controllable time data, and maintenance resource load balancing data. By constructing a multi-dimensional decision matrix, the risk level, maintenance capacity constraints, and resource availability under each working condition are comprehensively evaluated. The optimal maintenance timing and maintenance intensity are determined by fuzzy logic reasoning method. A dynamic adjustment mechanism for strategies is established to cope with changes in working conditions. Finally, maintenance controllable strategy adjustment data that takes into account both safety and economy is generated.

[0039] Example 2 This application also discloses a big data-driven engine maintenance management system.

[0040] Reference Figure 2 A big data-driven engine maintenance management system includes: The data acquisition module is used to acquire engine operation log data, extract operating environment data and engine component status data from the engine operation log data, and obtain operating environment data and engine component status data respectively. Based on the operating environment data, the module performs dynamic wear mapping of components between different operating conditions on the engine component status data to obtain wear data of components with different operating conditions. The analysis module is used to calculate the wear rate between different working conditions based on the wear data of components with different working conditions, obtain the wear rate of different working conditions, perform stress increment transmission analysis between component connection points based on the wear rate of different working conditions, obtain stress increment transmission data of connection points, and identify the critical failure interval of connection points based on the stress increment transmission data of connection points, thereby obtaining the critical failure interval of connection points. The adjustment module is used to adjust the controllable maintenance strategy between different working conditions based on the wear data of components with different working conditions according to the critical range of connection point failure, and to obtain normalized data for maintenance strategy adjustment. The management module is used to learn the maintenance control logic from the critical interval of connection point failure and the normalized data of maintenance strategy adjustment to obtain maintenance control logic data. Based on the maintenance control logic data, the module designs automated logic firmware to obtain maintenance control logic firmware. The maintenance control logic firmware is then embedded into the engine monitoring terminal to perform engine maintenance management.

[0041] The above content is merely an example and illustration of the concept of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described or use similar methods to replace them, as long as they do not deviate from the concept of the invention, they should all fall within the protection scope of the present invention.

[0042] In the description of this specification, references to terms such as "an embodiment," "example," "specific example," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0043] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to any specific implementation. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention.

Claims

1. A big data-driven engine maintenance management method, characterized in that, Includes the following steps: Step S1: Obtain engine operation log data, extract operating environment data and engine component status data from the engine operation log data to obtain operating environment data and engine component status data respectively, and perform dynamic wear mapping of components between different operating conditions based on the operating environment data to obtain wear data of components with different operating conditions. Step S2: Calculate the wear rate between different working conditions for the wear data of components with different working conditions to obtain the wear rate of different working conditions. Perform stress increment transmission analysis between component connection points based on the wear rate of different working conditions to obtain the stress increment transmission data of connection points. Identify the critical failure interval of connection points based on the stress increment transmission data of connection points to obtain the critical failure interval of connection points. Step S3: Based on the critical failure range of the connection point, adjust the controllable maintenance strategy for different working conditions of the wear data of components with different working conditions to obtain normalized data of maintenance strategy adjustment; Step S4: Perform maintenance control logic learning on the critical interval of connection point failure and the normalized data of maintenance strategy adjustment to obtain maintenance control logic data. Based on the maintenance control logic data, design automated logic firmware to obtain maintenance control logic firmware. Embed the maintenance control logic firmware into the engine monitoring terminal to perform engine maintenance management.

2. The big data-driven engine maintenance management method according to claim 1, characterized in that, Step S1 includes the following steps: Step S11: Obtain engine operation log data, and perform data cleaning and outlier processing on the engine operation log data to obtain cleaned operation log data; Step S12: Extract operating environment data and engine component status data from the cleaned operation log data respectively to obtain operating environment data and engine component status data. Step S13: Perform time-series alignment and interpolation on the engine component status data to obtain missing component status data; Step S14: Based on the working environment data, perform dynamic wear mapping of components under different working conditions to fill in missing component status data and obtain wear data of components with different working conditions.

3. The big data-driven engine maintenance management method according to claim 1, characterized in that, Step S2 includes the following steps: Step S21: Calculate the instantaneous wear rate between different working conditions for the wear data of components with different working conditions, and obtain the wear rate with different working conditions; Step S22: Simulate the instantaneous thermal stress response of the component based on the wear rate difference under working conditions to obtain the instantaneous thermal stress response data of the component; Step S23: Obtain engine design data, which includes component material properties, connection structure geometric parameters, and thermal conductivity coefficient; Step S24: Based on the wear rate under different operating conditions and the instantaneous thermal stress response data of the components, perform stress increment transmission analysis between component connection points on the engine design data to obtain stress increment transmission data at the connection points; Step S25: Identify stress superposition connection points by analyzing the stress increment transmission data at the connection points to obtain stress superposition connection points; Step S26: Based on the stress increment transmission data of the connection point and the engine design data, identify the critical failure interval of the connection point for stress superposition and obtain the critical failure interval of the connection point.

4. The big data-driven engine maintenance management method according to claim 3, characterized in that, Step S24 includes the following steps: Step S241: Perform topology connection structure analysis on the engine design data to obtain engine topology connection structure data; Step S242: Perform thermal stress flow guidance analysis on the engine topology connection structure data based on the instantaneous thermal stress response data of the components to obtain the thermal stress flow guidance data of the connection structure; Step S243: Analyze the temperature difference distribution at the connection point based on the thermal stress flow guidance data of the connection structure to obtain the temperature difference distribution data at the connection point; Step S244: Calculate the local connection point temperature difference increment series based on the wear rate of the working condition difference to obtain the local connection point temperature difference increment series; Step S245: Based on the local connection point temperature difference increment series and connection point temperature difference distribution data, perform stress increment transmission analysis between component connection points on the engine topology connection structure data to obtain connection point stress increment transmission data.

5. The big data-driven engine maintenance management method according to claim 3, characterized in that, Step S26 includes the following steps: Step S261: Analyze the material properties of the component connection lines in the engine design data to obtain the material property data of the component connection lines; Step S262: Calculate the stress vector of the stress superimposed connection point based on the stress increment transmission data of the connection point to obtain the stress vector data of the superimposed connection point; Step S263: Based on the stress vector data of the superimposed connection point, perform thermal expansion plastic strain fitting on the material property data of the component connection line to obtain thermal expansion plastic strain fitting data; Step S264: Calculate the fatigue tolerance of the superimposed connection point by performing stress vector data on the superimposed connection point to obtain fatigue tolerance data of the superimposed connection point; Step S265: Based on the thermal expansion plastic strain fitting data and the fatigue tolerance data of the superimposed points, identify the critical interval of connection point failure for the stress superimposed connection points, and obtain the critical interval of connection point failure.

6. The big data-driven engine maintenance management method according to claim 5, characterized in that, Step S263 includes the following steps: Step S2631: Extract the heat capacity of the component connection circuit material from the material property data to obtain the heat capacity of the circuit material; Step S2632: Calculate the average difference of thermal energy in the stress direction on the stress vector data of the superimposed connection point to obtain the average difference of thermal energy in the stress direction; Step S2633: Simulate the plastic yield limit of the heat capacity of the line material based on the average difference of thermal energy in the stress direction, and obtain the plastic yield data of the material with average difference of thermal energy. Step S2634: Based on the superimposed connection point stress vector data, perform plastic anisotropy analysis on the thermal energy uniformity material plastic yield data to obtain the material plastic yield anisotropy data; Step S2635: Based on the thermal energy uniformity difference material plastic yield data, the material plastic yield anisotropy data, and the stress direction thermal energy uniformity difference data, perform thermal expansion plastic strain fitting to obtain thermal expansion plastic strain fitting data.

7. The big data-driven engine maintenance management method according to claim 5, characterized in that, Step S264 includes the following steps: Step S2641: Analyze the change in heat flux density at the connection point by performing stress vector data at the superimposed connection point to obtain the change data of heat flux density at the connection point; Step S2642: Perform density increment trend analysis on the heat flux density change data at the connection point to obtain heat flux density increment trend data; Step S2643: Based on the heat flux density increment trend data, perform piecewise mapping of the fatigue gradient fluctuation data at the connection point to obtain the fatigue gradient piecewise fluctuation data. Step S2644: Perform fatigue limit numerical calculation on the fatigue gradient segmented fluctuation data to obtain the fatigue limit value; Step S2645: Based on the fatigue limit value and the fatigue gradient piecewise fluctuation data, calculate the fatigue tolerance of the superimposed connection point stress vector data to obtain the fatigue tolerance data of the superimposed point.

8. The big data-driven engine maintenance management method according to claim 1, characterized in that, Step S3 includes the following steps: Step S31: Analyze the lower limit safe operating threshold between different operating conditions based on the wear data of components with different operating conditions, and obtain the lower limit safe operating threshold between different operating conditions; Step S32: Based on the critical interval of connection point failure, adjust the lower limit safe operation threshold between different road conditions to obtain maintenance controllable strategy adjustment data. Step S33: Normalize the maintenance controllable strategy adjustment data to obtain normalized maintenance strategy adjustment data.

9. The big data-driven engine maintenance management method according to claim 8, characterized in that, Step S32 includes the following steps: Step S321: Based on the critical interval of connection point failure, calculate the upper limit of instantaneous maintenance capacity under different working conditions for the lower limit safe operation threshold of different road conditions, and obtain the upper limit data of instantaneous maintenance capacity; Step S322: Based on the critical interval of connection point failure, analyze the controllable maintenance limit time under different working conditions of the instantaneous maintenance load limit data to obtain the controllable maintenance limit time data; Step S323: Perform maintenance resource load balancing processing based on instantaneous maintenance capacity limit data and maintenance capacity controllable duration data to obtain maintenance resource load balancing data; Step S324: Based on the instantaneous maintenance capacity limit data, the maintenance limit continuous controllable time data, and the maintenance resource load balance data, adjust the maintenance controllable strategy for different working conditions according to the lower limit safe operation threshold for different road conditions, and obtain maintenance controllable strategy adjustment data.

10. A big data-driven engine maintenance management system, applied to the big data-driven engine maintenance management method described in any one of claims 1-9, characterized in that, include: The data acquisition module is used to acquire engine operation log data, extract operating environment data and engine component status data from the engine operation log data, and obtain operating environment data and engine component status data respectively. Based on the operating environment data, the module performs dynamic wear mapping of components between different operating conditions on the engine component status data to obtain wear data of components with different operating conditions. The analysis module is used to calculate the wear rate between different working conditions based on the wear data of components with different working conditions, obtain the wear rate of different working conditions, perform stress increment transmission analysis between component connection points based on the wear rate of different working conditions, obtain stress increment transmission data of connection points, and identify the critical failure interval of connection points based on the stress increment transmission data of connection points, thereby obtaining the critical failure interval of connection points. The adjustment module is used to adjust the controllable maintenance strategy between different working conditions based on the wear data of components with different working conditions according to the critical range of connection point failure, and to obtain normalized data for maintenance strategy adjustment. The management module is used to learn the maintenance control logic from the critical interval of connection point failure and the normalized data of maintenance strategy adjustment to obtain maintenance control logic data. Based on the maintenance control logic data, the module designs automated logic firmware to obtain maintenance control logic firmware. The maintenance control logic firmware is then embedded into the engine monitoring terminal to perform engine maintenance management.

Images