A method, device and medium for predicting the service life of a jet pre-cooling compressor blade

Abstract

The application discloses a kind of jet precooling compressor blade life prediction method, equipment and medium, it is related to blade health management technical field, including according to residual life and damage state data, using digital twinborn method, the failure risk of blade health state and efficiency loss are weighed online, and jet precooling intervention instruction is generated by conservative exploration;Jet precooling intervention instruction is executed, and cooling parameter is dynamically adjusted, and the measured temperature and strain information of sensor acquisition after cooling are obtained;Based on measured temperature and strain information, blade material parameters are corrected, and the corrected blade material parameters are updated, and the updated life prediction result is obtained.The application is analyzed by using coupling model to the working condition of compressor last stage blade, the joint identification of multi-physical field damage mechanism is realized, the residual life and damage state data are accurately output, and high confidence basis is provided for life management.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of blade health management, and in particular to a jet pre-cooling compressor blade life prediction method, device and medium. BACKGROUND

[0002] In the field of aero-engines and gas turbines, the last stage blade of the compressor is long-term under the coupling action of high temperature, high speed and complex aerodynamic load, and life prediction is crucial to ensure equipment reliability. The existing technology usually combines fluid-solid coupling simulation and fatigue damage model to analyze the thermal response of the blade under the condition of jet pre-cooling, and estimates the remaining life based on the Miner linear cumulative rule, and introduces digital simulation means to assist evaluation.

[0003] However, the conventional method usually uses a single damage mechanism model, which is difficult to accurately depict the nonlinear influence of the interaction of creep, fatigue and oxidation on life under high temperature environment; at the same time, the existing life prediction is mostly offline static evaluation, lacking online closed-loop feedback mechanism with jet control compressor blade, and unable to realize dynamic correction and active intervention based on measured response. SUMMARY

[0004] In view of the above existing problems, the present application is proposed.

[0005] Therefore, the present application provides a jet pre-cooling compressor blade life prediction method to solve the problem that life prediction cannot be real-time calibrated and iterated.

[0006] To solve the above technical problems, the present application provides the following technical solutions:

[0007] In a first aspect, the present application provides a jet pre-cooling compressor blade life prediction method, which comprises using a coupling model to analyze the working condition of the last stage blade of the compressor, and obtaining residual life and damage state data; according to the residual life and damage state data, using a digital twin method to online weigh the failure risk and efficiency loss of the blade health state, generating a jet pre-cooling intervention instruction through conservative exploration; executing the jet pre-cooling intervention instruction and dynamically adjusting the cooling parameters to obtain measured temperature and strain information collected by the sensor after cooling; based on the measured temperature and strain information, correcting the blade material parameters, and updating the corrected blade material parameters to obtain updated life prediction results.

[0008] As a preferred scheme of the jet pre-cooling compressor blade life prediction method of the present application, wherein: the coupling model, the specific construction steps are as follows,

[0009] The multi-source blade data is processed by time and space alignment and unification to obtain an observation set and a coupling factor;

[0010] Based on the observation set and the coupling factor, self-consistent calibration is performed to obtain a coupling state sequence and an intermediate state summary;

[0011] According to the coupling state sequence and the intermediate state summary, comparative evaluation and aggregate evaluation are performed to obtain a comprehensive evaluation quantity and a ranking result, and a coupling model is generated.

[0012] As a preferred scheme of the jet pre-cooling compressor blade life prediction method, wherein: the coupling model is used to analyze the working condition of the compressor final stage blade to obtain residual life and damage state data, and the specific steps are as follows,

[0013] Based on the coupling model, multi-source working condition data is processed to obtain a coupling factor table;

[0014] By analyzing the temperature field and the stress-strain field jointly based on the coupling factor table, a position response vector is obtained;

[0015] By using the life and damage evaluation method, the position response vector is subjected to damage accumulation identification to obtain residual life and damage state data.

[0016] As a preferred scheme of the jet pre-cooling compressor blade life prediction method, wherein: the digital twin method has the following specific construction steps,

[0017] Based on the working condition data and the damage data, comparative evaluation and aggregate evaluation are integrated to obtain a mapping baseline and an object digital image;

[0018] According to the mapping baseline and the object digital image, two-way synchronization and deviation correction are performed through self-consistent calibration to obtain an online simulation channel and a calibrated state set;

[0019] According to the online simulation channel and the calibrated state set, health assessment, prediction simulation and intervention deduction are closed-loop operated to obtain driving instructions, early warning lists and interpretable points, and a digital twin method is generated.

[0020] As a preferred scheme of the jet pre-cooling compressor blade life prediction method, wherein: according to the residual life and the damage state data, a digital twin method is used to online weigh the failure risk and the efficiency loss of the blade health state, and a jet pre-cooling intervention instruction is generated through conservative exploration, and the specific steps are as follows,

[0021] Based on the residual life and the damage state data, the current health state of the compressor final stage blade is comprehensively evaluated to obtain a quantitative relationship between the blade failure risk level and the operating efficiency loss degree;

[0022] According to the quantitative relationship, the cooling intensity and the jet timing of the jet pre-cooling are optimized by using a conservative exploration strategy to generate a jet pre-cooling intervention instruction.

[0023] As a preferred scheme of the jet pre-cooling compressor blade life prediction method, the jet pre-cooling intervention instruction is executed, and the cooling parameters are dynamically adjusted to obtain the measured temperature and strain information after cooling.

[0024] According to the jet pre-cooling intervention instruction, the jet opening, pulse frequency and nozzle selection are issued and synchronized to form a preliminary cooling response signal.

[0025] Through the preliminary cooling response signal, the mass flow, supply temperature and duty cycle are adjusted in a closed loop using self-consistent calibration to obtain a cooling parameter sequence.

[0026] According to the cooling parameter sequence, time sequence sampling and integrity verification are performed to obtain the measured temperature and strain information after cooling.

[0027] As a preferred scheme of the jet pre-cooling compressor blade life prediction method, the cooling parameter sequence refers to a set of adjustable control quantities for controlling the intensity and distribution of jet pre-cooling, including mass flow, supply temperature, jet pressure and intensity, nozzle selection, jet timing, pulse frequency and duty cycle.

[0028] As a preferred scheme of the jet pre-cooling compressor blade life prediction method, the cooling parameter sequence refers to a set of adjustable control quantities for controlling the intensity and distribution of jet pre-cooling, including mass flow, supply temperature, jet pressure and intensity, nozzle selection, jet timing, pulse frequency and duty cycle.

[0029] Based on the measured temperature and strain information, the multi-source measurement point data is cleaned, time-aligned and steady-state section identified using a consistent method to obtain baseline sequence and load condition data.

[0030] According to the baseline sequence and load condition data, the elastic, strength and creep material parameters are identified and corrected in amplitude using self-consistent calibration to obtain the corrected blade material parameter set.

[0031] According to the corrected blade material parameter set, the cyclic damage and high temperature degradation are co-optimized and continuously corrected to obtain the updated life prediction result.

[0032] In a second aspect, the present application provides a computer device comprising a memory and a processor, the memory storing a computer program, wherein the computer program is executed by the processor to implement any step of the jet pre-cooling compressor blade life prediction method according to the first aspect of the present application.

[0033] In a third aspect, the present application provides a computer readable storage medium having stored thereon a computer program, wherein the computer program, when executed by a processor, implements any step of the jet precooling compressor blade life prediction method according to the first aspect of the present application.

[0034] The present application has the following beneficial effects: by using the coupling model to analyze the working condition of the last stage blade of the compressor, the joint identification of the multi-physical field damage mechanism is realized, the residual life and damage state data are accurately output, and high confidence basis is provided for life management; by using the digital twin method to online weigh the failure risk and efficiency loss and generate jet precooling intervention instructions, the active cooling regulation and control driven by prediction is realized, and the effectiveness of the synergistic optimization of blade life extension and operation efficiency is achieved. BRIEF DESCRIPTION OF DRAWINGS

[0035] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0036] Fig. 1 The flowchart of the jet precooling compressor blade life prediction method.

[0037] Fig. 2 The flowchart of the coupling model processing chain.

[0038] Fig. 3 The flowchart of the digital twin method closed loop.

[0039] Fig. 4 The flowchart of the execution and parameter update chain. DETAILED DESCRIPTION

[0040] In order to make the above-mentioned purposes, features and advantages of the present application more apparent and easy to understand, the specific embodiments of the present application will be described in detail below with reference to the drawings of the specification.

[0041] In the following description, many specific details are set forth in order to provide a thorough understanding of the present application, but the present application can also be implemented in other ways different from those described herein, and those skilled in the art can make similar generalizations without departing from the connotation of the present application, therefore the present application is not limited to the specific embodiments disclosed below.

[0042] Secondly, the "one embodiment" or "embodiment" referred to herein means that the specific features, structures or characteristics can be included in at least one implementation of the present application. "In one embodiment" appearing in different places in the specification does not mean the same embodiment, nor is it an independent or alternative embodiment that excludes other embodiments.

[0043] Referring to Figs. 1-4 For an embodiment of the present application, the embodiment provides a jet pre-cooling compressor blade life prediction method, comprising the following steps:

[0044] S1: using a coupling model to analyze the working condition of the last stage blade of the compressor to obtain residual life and damage state data.

[0045] S1.1: performing spatio-temporal alignment and consistent processing on multi-source blade data to obtain an observation set and a coupling factor.

[0046] Specifically, the multi-source blade data is composed of joint collection and simulation output of working conditions and structural responses.

[0047] Further, by synchronously acquiring temperature fields, strain responses and operating condition parameters from the infrared thermal imager, blade tip timing sensor and engine control deployed on the last stage blade of the compressor, the multi-source data of different sampling frequencies are unified to the same time reference by consistent processing, and the blade data is mapped to a unified numerical space by denoising and consistent processing, forming a structured observation set and a coupling factor.

[0048] It should be noted that denoising and consistent processing refers to based on spatio-temporal aligned blade data, reorganizing temperature and strain sequences by sampling, completing noise suppression according to steady state section identification, unifying time markers and data formats through integrity verification, and outputting observation sets and coupling factors on common time reference and unified field definition according to consistent method.

[0049] S1.2: based on the observation set and the coupling factor, self-consistent calibration is performed to obtain a coupling state sequence and an intermediate state abstract.

[0050] Further, the multi-source working condition data in the observation set and the coupling factor are corrected by self-consistent calibration to correct the coupling response of the temperature field and the stress-strain field, generate a coupling state sequence representing the multi-field coupling process of the blade, and extract feature information to form an intermediate state abstract.

[0051] It should be noted that self-consistent calibration refers to the process of iterative alignment and common correction of multi-source working condition data, coupling model and digital twin simulation, with the goal of minimizing residual error, synchronously correcting multi-source working condition data and coupling factor, completing denoising, time series consistent and anomaly suppression, obtaining a coupling state sequence and an intermediate state abstract.

[0052] Preferably, during the execution of the jet pre-cooling intervention process, the mass flow, the supply air temperature and the duty cycle are closed-loop adjusted by self-consistent calibration to obtain a cooling parameter sequence; in the life prediction update stage, according to the baseline sequence and the load working condition data, the elastic modulus, the strength and the creep material parameters are identified and corrected in amplitude by self-consistent calibration based on the baseline sequence and the load working condition data, to obtain a corrected blade material parameter set.

[0053] S1.3: According to the coupling state sequence and the intermediate state summary, comparative evaluation and aggregation evaluation are performed to obtain a comprehensive evaluation quantity and a ranking result, and a coupling model is generated.

[0054] Further, the state characteristics of each time in the coupling state sequence are compared with the intermediate state summary, and the consistency processing is performed according to the results of the joint analysis of the temperature field and the stress and strain field. The multi-source evaluation results are fused by using the aggregation evaluation method to form a comprehensive evaluation quantity reflecting the overall degradation trend, and the damage data under different working conditions are ranked based on the comprehensive evaluation quantity to generate the coupling model.

[0055] It should be noted that the aggregation evaluation method refers to the unified quantification of multi-source working condition data and multi-source evaluation results by weighted fusion and uncertainty calibration, and the output of a comprehensive evaluation quantity and a ranking support risk judgment and intervention decision.

[0056] Specifically, the damage data refers to the damage accumulation identification of the position response vector based on the joint analysis results of the temperature field and the stress and strain field, which is used to represent the damage degree, type and distribution of the last stage blade of the compressor.

[0057] In the coupling model construction process, the multi-source blade data is time and space aligned and consistency processed to obtain an observation set and a coupling factor. Based on the observation set and the coupling factor, self-consistent calibration is performed to obtain a coupling state sequence and an intermediate state summary. According to the coupling state sequence and the intermediate state summary, comparative evaluation and aggregation evaluation are performed to obtain a comprehensive evaluation quantity and a ranking result, and a coupling model is generated.

[0058] S1.4: Based on the coupling model, the multi-source working condition data is denoised and consistency processed to obtain a coupling factor table.

[0059] Further, by denoising to eliminate high-frequency noise and abnormal fluctuations, and then using consistency processing to map the blade data to a unified field definition and numerical range consistency, the processed data is grouped and aggregated according to the working condition characteristics and physical relevance to form a coupling factor table.

[0060] S1.5: The position response vector is obtained by performing joint analysis of the temperature field and the stress and strain field on the coupling factor table.

[0061] Further, based on the multi-source working condition data contained in the coupling factor table after denoising and unification processing, the thermal response of the last stage blade of the compressor under the condition of jet pre-cooling (the comprehensive performance of the temperature field, heat flux density, thermal stress and deformation of the last stage blade of the compressor and the surrounding flow field due to temperature and pressure changes when jet pre-cooling is implemented) is jointly analyzed by using the coupling model, and the transient temperature distribution of each point on the blade surface and inside and the corresponding stress and strain distribution are obtained. The thermal-mechanical coupling response of the stress and strain distribution is jointly analyzed to form a position response vector.

[0062] S1.6: Through the life and damage assessment method, the position response vector is subjected to damage accumulation identification, and residual life and damage state data are obtained.

[0063] Further, based on the position response vector (the position response vector is derived from the joint analysis of the temperature field and the stress and strain field), the life and damage assessment method is used to segmentally accumulate the coupling effect of blade damage and temperature strain, and a damage evolution sequence is obtained. According to the damage evolution sequence, the health consumption of the last stage blade of the compressor is quantified through damage accumulation identification, and residual life estimation and damage characteristic labeling are generated. Through residual life estimation and damage characteristic labeling, spatial distribution and type are summarized, mapped and state-archived to obtain residual life and damage state data.

[0064] It should be noted that the life and damage assessment method refers to the joint quantification of cyclic damage and high-temperature degradation based on the position response vector and the working condition history through segmental accumulation and time sequence tracking (joint quantification is to fuse the corrected blade material parameter set with the temperature field and stress and strain field data under the current working condition, use the comparison evaluation and aggregation evaluation mechanism in the coupling model to simultaneously weight the cyclic damage and high-temperature degradation effects, output a unified damage evolution path and residual life estimation, and realize the consistency quantification of life state under the influence of multiple physical fields). According to the coupling changes of the temperature field and the stress and strain field, the damage evolution sequence and the residual life estimation are obtained, and through spatial distribution mapping and type archiving, the residual life and damage state data are obtained.

[0065] Specifically, the comparison evaluation mechanism refers to the inspection of the stress and strain field data under different working conditions under a unified baseline to identify the sources of deviation and the superior and inferior schemes.

[0066] The aggregation evaluation mechanism refers to the uncertainty calibration of the multi-source working condition data and the stress and strain field data to generate a single comprehensive evaluation quantity and ranking for decision-making.

[0067] S2: According to the residual life and damage state data, a digital twin method is used to online weigh the failure risk and efficiency loss of the blade health state, and jet pre-cooling intervention instructions are generated through conservative exploration.

[0068] S2.1: Based on the working condition data and damage data, comparative evaluation and integrated aggregation evaluation are carried out to obtain the mapping baseline and the object digital portrait.

[0069] Further, based on the working condition data and damage data, the comparative evaluation method is used to analyze the differences of the compressor last stage blade under different operating conditions (referring to the different states of the compressor in combination of speed, load, inlet temperature and pressure, humidity, flow and flow distortion, and pre-cooling parameters external and control variables), and the common law under multiple working conditions is comprehensively summarized by combining the aggregation evaluation method. The mapping baseline capable of representing the current state and degradation trajectory of the blade is formed after integration, and the mapping baseline and the object digital portrait are obtained.

[0070] Specifically, the comparative evaluation method refers to the difference analysis of the temperature field and stress-strain field response under different working conditions and time sequences based on the observation set and coupling factor, as well as the coupling state sequence and intermediate state abstract, to generate a comparison image and a reference sequence.

[0071] The aggregation evaluation method refers to the consistent processing of the benchmarked multi-source working condition data by combining the difference measure from the comparative evaluation method, to generate a comprehensive evaluation quantity and a sorting result.

[0072] Preferably, the difference analysis is based on the observation set and coupling factor, as well as the coupling state sequence and intermediate state abstract, and the relative increase and decrease of the temperature field and stress-strain field under different working conditions and time sequences are analyzed by the comparative evaluation method to generate a difference measure and a comparison image, which provides analysis for priority judgment and comprehensive evaluation quantity calculation for the aggregation evaluation method.

[0073] S2.2: According to the mapping baseline and the object digital portrait, two-way synchronization and deviation correction are carried out by self-consistent calibration to obtain the online simulation channel and the calibrated state set.

[0074] Further, the actual operating state reflected by the mapping baseline and the object digital portrait is two-way synchronized and deviation corrected, and the embedded physical consistency constraint condition is iteratively modified by self-consistent calibration, so that the simulation state in the digital space and the measured response in the physical space are kept synchronized in time sequence.

[0075] Preferably, during the calibration process, the simulation boundary conditions and material response parameters are continuously adjusted by self-consistent calibration to eliminate cumulative errors, generate an online simulation channel that reflects the current health status of the compressor last stage blade in real time, and output a calibrated state set that is highly consistent with the physical entity, to obtain the online simulation channel and the calibrated state set.

[0076] It should be pointed out that the bidirectional synchronization and deviation correction refers to calibrating the online simulation channel and the observation set based on the mapping baseline and the object digital image, correcting the deviation sources item by item according to the self-consistent calibration, and updating the cooling parameter sequence of the jet pre-cooling intervention and the measured temperature and strain information after cooling in both directions, to obtain the online simulation channel and the calibrated state set.

[0077] S2.3: According to the online simulation channel and the calibrated state set, the health assessment, prediction simulation and intervention deduction are closed-loop run to obtain the driving instructions, early warning list and explainable points, and the digital twin method is generated.

[0078] Further, the working condition data and damage state data of the compressor last stage blade are continuously received by the online simulation channel, and the current health status of the blade is dynamically health assessed in combination with the calibrated state set, and the life evolution trend under different jet pre-cooling strategies is closed-loop run through the prediction simulation, and the trade-off between failure risk and efficiency loss is made (the trade-off is made according to the residual life and damage state data to comprehensively evaluate the current health status of the compressor last stage blade, to obtain the quantitative relationship between the blade failure risk level and the operating efficiency loss degree, and according to the quantitative relationship, the cooling intensity and injection timing of the jet pre-cooling are coordinated and optimized by using the conservative exploration strategy to minimize the adverse effects on aerodynamic efficiency while ensuring the safety of the blade structure, to achieve a reasonable balance between health risk and operating performance), to obtain the jet pre-cooling intervention instructions as the driving instructions, to obtain the early warning list and the explainable points reflecting the basis for decision-making, and to generate the digital twin method.

[0079] S2.4: Based on the residual life and damage state data, the current health status of the compressor last stage blade is comprehensively evaluated to obtain the quantitative relationship between the blade failure risk level and the operating efficiency loss degree.

[0080] Further, the residual life and damage state data are input into the digital twin method, and the mapping baseline and the object digital image obtained by the digital twin method are used to jointly analyze the structural integrity and thermal performance degradation trend of the compressor last stage blade under the current working condition (based on the coupling model, the multi-source working condition data is denoised and unified to form a coupling factor table, the measured temperature and strain are aligned with the online simulation channel under the self-consistent calibration, the temperature field and stress-strain field joint analysis is carried out to obtain the position response vector, and the degradation trend of the structural integrity and thermal performance is identified through the life and damage evaluation method), and the comprehensive evaluation data is obtained through the online simulation channel for health assessment, prediction simulation and intervention deduction, and the quantitative relationship between the blade failure risk level and the operating efficiency loss degree is analyzed through the comprehensive evaluation data.

[0081] Specifically, the comprehensive evaluation data refers to the comprehensive evaluation quantity and the ranking result formed by aggregating the evaluation quantity generated by the comparative evaluation method based on the observation set and the coupling factor, combining the coupling state sequence and the intermediate state summary, and merging the difference measure generated by the comparative evaluation method, and the leaf data set used for judgment and decision-making. The comprehensive evaluation quantity is obtained by combining the risk-related measure and the efficiency-related measure.

[0082] S2.5: According to the quantitative relationship, the cooling intensity and the jetting timing of the jet pre-cooling are optimized by using a conservative exploration strategy, and a jet pre-cooling intervention instruction is generated.

[0083] Further, based on the online simulation channel and the calibrated state set constructed by the digital twin method, based on the quantitative relationship between the blade failure risk level and the operating efficiency loss degree, the cooling intensity and the jetting timing of the jet pre-cooling are iteratively analyzed by using a conservative exploration strategy, and the cooling parameter combination that minimally affects the operating efficiency is selected under the premise of ensuring that the blade damage is not aggravated. Through the closed-loop operation process of health assessment, prediction simulation and intervention deduction in the digital twin method, a jet pre-cooling intervention instruction that meets the conservative exploration constraint condition is output.

[0084] It should be noted that the conservative exploration strategy refers to the risk and efficiency trade-off of the cooling intensity and the jetting timing based on the remaining life and damage state data and the online evaluation results of the digital twin method (the small amplitude exploration and directional adjustment is performed by perturbing the cooling parameters of mass flow, jet timing or pulse frequency in a small step within the health risk boundary determined by the current remaining life and damage state data, observing the response change trend of the failure risk level and the efficiency loss degree output by the digital twin method, and determining the optimization direction and gradually converging to the jet pre-cooling intervention instruction that maximizes the life gain and minimizes the operating efficiency loss), and gradually converging to a stable scheme in the safe region through closed-loop adjustment and consistent processing, generating a jet pre-cooling intervention instruction and providing constraints and update basis for the cooling parameter sequence.

[0085] S3: Execute the jet pre-cooling intervention instruction and dynamically adjust the cooling parameters to obtain the measured temperature and strain information collected by the sensor after cooling.

[0086] S3.1: According to the jet pre-cooling intervention instruction, the jetting opening, pulse frequency and jet selection are issued and synchronized to form a preliminary cooling response signal.

[0087] Further, the jet pre-cooling intervention instruction contains the jet opening signal, pulse frequency parameter and nozzle selection instruction information (the specific instruction information refers to the control content, including jet opening and stopping, nozzle selection, cooling intensity, jet timing, pulse frequency and duty cycle and synchronization requirements) in the jet pre-cooling intervention instruction. The jet pre-cooling intervention instruction is synchronously transmitted to the jet execution mechanism through the control bus (the time stamp of the data and the instruction is derived from the unified time reference) to enable multiple nozzles to start the jetting action in accordance with the timing and frequency of the instruction information, so that a cooling liquid distribution consistent with the instruction is formed in the flow passage of the last stage blade of the compressor. The flow field disturbance and temperature change generated in the jetting process are preliminarily sensed by the sensor to form a preliminary cooling response signal.

[0088] S3.2: Through the preliminary cooling response signal, the self-consistent calibration is used to close-loop adjust the mass flow, gas supply temperature and duty cycle to obtain a cooling parameter sequence.

[0089] Further, after the jetting action is triggered according to the jet pre-cooling intervention instruction, the preliminary cooling response signal fed back by the infrared thermal imaging and the blade tip timing sensor (the infrared thermal imaging is used to non-contact acquire the blade surface temperature field and thermal spot distribution, and the blade tip timing sensor is used to record the blade tip arrival time deviation to represent the rotor blade vibration deflection and stress state) is collected. The preliminary cooling response signal reflects the blade surface temperature change trend and structural strain response characteristics under the current cooling action. The self-consistent calibration method is used to compare the preliminary cooling response signal with the target temperature distribution and strain response in the mapping baseline, and the mass flow, gas supply temperature and duty cycle in the jet pre-cooling are iteratively adjusted, so that the cooling action gradually approaches the optimal intervention state. Each adjustment is based on the deviation correction of the previous response result until a cooling parameter sequence that is continuous in time sequence and consistent in parameters is formed. The cooling parameter sequence completely represents the dynamic evolution path of each adjustable control quantity in the jet pre-cooling process.

[0090] S3.3: According to the cooling parameter sequence, the timing sampling and integrity verification are performed to obtain the measured temperature and strain information after cooling.

[0091] Further, according to the cooling parameter sequence, the output signals of the infrared thermal imager and the blade tip timing sensor are sequentially sampled in time sequence to obtain temperature and strain time sequence data strictly aligned with the cooling parameter sequence. The integrity verification is performed according to the unification method and time stamp alignment to eliminate the missing segments caused by instantaneous failure of the sensor and communication interruption, retain the continuous and consistent cooling response segments, and obtain the measured temperature and strain information after cooling.

[0092] It should be noted that the timing sampling refers to continuously acquiring the temperature and strain sequence of multiple source measuring points based on the unified time reference and completing the time synchronization to form a sampling sequence that can be used for subsequent evaluation.

[0093] The integrity verification refers to checking the time stamp continuity of the sampling sequence, the corresponding relationship of the multi-source measuring points and the data coverage based on a unified time reference and using a consistent method.

[0094] S4: Based on the measured temperature and strain information, the blade material parameters are corrected, and the corrected blade material parameters are updated to obtain the updated life prediction results.

[0095] S4.1: The cooling parameter sequence refers to a set of adjustable control quantities for controlling the strength and distribution of the jet pre-cooling, including mass flow, supply air temperature, jet pressure and intensity, nozzle selection, jet timing, pulse frequency and duty cycle.

[0096] Preferably, the jet pre-cooling intervention instruction is issued and synchronized to generate a preliminary cooling response signal, and then the mass flow, supply air temperature and duty cycle are adjusted in a closed loop using self-consistent calibration to obtain the cooling parameter sequence based on the preliminary cooling response signal. The cooling parameter sequence is used for timing sampling and integrity verification to obtain the measured temperature and strain information after cooling.

[0097] S4.2: Based on the measured temperature and strain information, the multi-source measuring point data is cleaned, time-aligned and steady-state section identified using a consistent method to obtain the baseline sequence and load condition data.

[0098] Further, the measured temperature and strain information from the infrared thermal imager and the tip timing sensor are subjected to outlier rejection and noise filtering to complete data cleaning; the cleaned measured temperature and strain information is time-stamped and aligned according to the compressor speed signal and sampling timing to ensure that the multi-source measuring point data is synchronized under a unified time reference; and the time series trend of the measured temperature and strain information is analyzed by a consistent method to identify the continuous time period in which the temperature and strain fluctuation amplitude is within a stable range as a steady-state section; the cleaned, time-aligned and steady-state section-containing measured temperature and strain information is integrated and used to modify the baseline sequence and compressor operating condition parameters to obtain the baseline sequence and load condition data (the load condition data is obtained from the speed, pressure, flow and jet parameters in the engine control).

[0099] It should be noted that the consistent method refers to cleaning and denoising, time and space alignment, dimension consistency processing and steady-state section identification of multi-source measuring point data and operating condition data, and forming a standardized observation set and baseline sequence in combination with integrity verification to eliminate format and scale inconsistencies caused by source differences, and providing a unified and reliable input for self-consistent calibration, coupling evaluation and life prediction.

[0100] Preferably, the purity of the measured temperature and strain information is improved by removing outliers and filtering noise to reduce the influence of random disturbances on quantization; the synchronization of multi-source measurement points under a unified time reference is achieved by timestamp alignment based on the compressor speed signal and the sampling time sequence, ensuring spatial correspondence and time sequence consistency; the time series trend of the measured temperature and strain information is analyzed by the uniformization method to highlight the continuous segments with clear physical meaning, repeatability and comparability, and to weaken the deviation caused by transient disturbances; the baseline sequence and load condition data for correction have higher stability and traceability, providing reliable input for self-consistent calibration, material parameter correction and life prediction, reducing error propagation and improving the convergence and effectiveness of closed-loop updating.

[0101] It should be noted that timestamp alignment refers to unifying the recording times of multi-source measurement points of infrared thermal imaging and blade tip timing sensors to the same time reference based on the compressor speed signal and the sampling time sequence, ensuring one-to-one correspondence and synchronization of blade data.

[0102] S4.3: According to the baseline sequence and load condition data, the deviation identification and amplitude correction of elastic, strength and creep material parameters are performed by self-consistent calibration, and the corrected blade material parameter set is obtained.

[0103] Further, the measured temperature and strain information corresponding to the steady-state section in the baseline sequence is time-matched with the load condition data, the deviation between the strain response output by the current coupled model under the same working condition and the measured strain is compared by the self-consistent calibration method, the values of the elastic, strength and creep material parameters are adjusted in reverse according to the deviation between the strain response and the measured strain, the correction direction and value of the material elastic modulus are back calculated by comparing the measured deformation with the coupled model predicted deformation, the output response of the coupled model is consistent with the actual mechanical response of the physical blade, the accuracy of life prediction is improved, and the corrected blade material parameter set is formed.

[0104] It should be noted that deviation identification and amplitude correction refers to: based on the measured temperature and strain information after timestamp alignment and the corresponding output of the online simulation channel, the response difference is quantified under the same working condition and time sequence by a comparative evaluation method, and a deviation distribution is formed; according to the deviation distribution, the elastic material parameters, strength material parameters and creep material parameters are corrected one by one by self-consistent calibration and updated back and forth; the consistency is inspected within the steady-state section by the uniformization method until the online simulation channel matches the measured temperature and strain information, and the corrected blade material parameter set is generated.

[0105] S4.4: According to the modified blade material parameter set, the cyclic damage and high temperature degradation are cooperatively optimized and continuously corrected to obtain an updated life prediction result.

[0106] Further, the modified blade material parameter set is input into the coupling model, and based on the temperature field and stress-strain field joint analysis results under the current working condition, the damage accumulation identification process of the position response vector is re-executed. According to the life and damage evaluation method, the fatigue damage caused by cyclic loading and the material degradation effect caused by high temperature environment are cooperatively optimized through comparative evaluation and aggregation evaluation of the coupling model, and the blade material of cyclic damage and high temperature degradation is continuously corrected, and the residual life and damage state data consistent with the current blade state are output as the updated life prediction result.

[0107] The advantages are as follows: based on the modified blade material parameter set and the temperature field and stress-strain field joint analysis results of the coupling model, the health quantification is realized through the damage accumulation identification of the position response vector, which is synchronized with the current blade state; according to the life and damage evaluation method and combined with the comparative evaluation and aggregation evaluation of the coupling model, the consistency of the judgment under the combined action of cyclic loading and high temperature degradation is improved; through the continuous correction of the blade material of cyclic damage and high temperature degradation to form an evaluation-intervention-feedback closed loop, the residual life and damage state data consistent with the running condition are output, and the timeliness and availability of the prediction are strengthened.

[0108] Specifically, the cyclic damage and high temperature degradation are caused by repeated stress-strain cycles due to speed variation, load fluctuation and vibration excitation, and the repeated tension and compression of the strain region caused by the combined action of thermal gradient and centrifugal field, which leads to crack initiation and propagation;

[0109] The high temperature degradation is caused by creep, oxidation and microstructure evolution under long-term high temperature, and the weakening of grain boundaries and phase boundaries leads to gradual decline of bearing capacity; thermal spots and uneven cooling increase the local temperature difference, producing additional load to accelerate the damage process of low cycle and high cycle; through mutual coupling, creep and oxidation reduce the crack tip resistance, and cyclic loading promotes damage accumulation, causing rapid consumption of life.

[0110] The embodiment also provides a computer device suitable for the jet pre-cooling compressor blade life prediction method, which comprises a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute the computer executable instructions to realize the jet pre-cooling compressor blade life prediction method proposed in the above embodiment.

[0111] The computer device can be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected by a system bus. The processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for running the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is configured to perform wired or wireless communication with an external terminal. The wireless communication can be achieved by WIFI, an operator network, NFC (Near Field Communication) or other technologies. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer overlaid on the display screen, or a key, trackball or touchpad arranged on the shell of the computer device, or an external keyboard, touchpad or mouse, etc.

[0112] The embodiment also provides a storage medium having a computer program stored thereon, the program being executed by a processor to implement the jet pre-cooling compressor blade life prediction method proposed in the above embodiment; and the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as a static random access memory (SRAM), an electrically erasable programmable read-only memory (EEPROM), an erasable programmable read-only memory (EPROM), a programmable read-only memory (PROM), a read-only memory (ROM), a magnetic memory, a flash memory, a magnetic disk or an optical disk.

[0113] To sum up, the present application achieves the joint identification of multi-physical field damage mechanisms by using a coupling model to analyze the working condition of the last stage blade of the compressor, accurately outputs residual life and damage state data, and provides high-confidence basis for life management; the failure risk and efficiency loss are online weighed by the digital twin method to generate jet pre-cooling intervention instructions, active cooling regulation and control driven by prediction are achieved, and the effectiveness of the synergistic optimization of blade life extension and operation efficiency is achieved.

[0114] It should be noted that the above examples are only used to illustrate the technical solutions of the present application but not limit the present application. Although the present application is described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or equivalently replaced, without departing from the spirit and scope of the technical solutions of the present application, which should be covered in the scope of the claims of the present application.

Claims

1. A method for predicting the lifespan of a jet-precooled compressor blade, characterized in that: include, Spatiotemporal alignment and unification processing of multi-source blade data were performed to obtain the observation set and coupling factor; Based on the observation set and coupling factor, self-consistent calibration is performed to obtain the coupled state sequence and intermediate state summary; The intermediate state summary is generated by correcting the coupling response of the temperature field and stress-strain field by combining the multi-source operating condition data and coupling factors in the measurement set through self-consistent calibration, generating a coupling state sequence characterizing the multi-field coupling process of the blade, and extracting feature information to form an intermediate state summary. Based on the coupled state sequence and intermediate state summary, comparative evaluation and aggregate evaluation are performed to obtain the comprehensive evaluation quantity and ranking results, and a coupled model is generated. A coupled model was used to analyze the operating conditions of the compressor's last-stage blades and obtain data on remaining life and damage status. Based on the coupling model, noise reduction and standardization processing are performed on multi-source operating condition data to obtain a coupling factor table; The coupling factor table is formed by denoising and unifying multi-source operating condition data, and then grouping and aggregating the processed data according to operating condition characteristics and physical correlation. By performing a joint analysis of the temperature field and stress-strain field using the coupling factor table, the position response vector is obtained. Damage accumulation is identified by analyzing the location response vector using lifetime and damage assessment methods to obtain remaining lifetime and damage status data. Based on remaining lifetime and damage status data, a digital twin method is used to weigh the failure risk and efficiency loss of blade health status online, and generate jet precooling intervention commands through a conservative exploration strategy. The conservative exploration strategy refers to using digital twin methods to weigh the failure risk and efficiency loss of the blade's health status online based on remaining lifetime and damage status data, and to make small-scale trials and directional adjustments to the cooling intensity and injection timing. The jet pre-cooling intervention command is executed, and the cooling parameters are dynamically adjusted to obtain the measured temperature and strain information after cooling collected by the sensor. Based on measured temperature and strain information, the blade material parameters are corrected, and the corrected blade material parameters are updated to obtain the updated life prediction results.

2. The method for predicting the lifespan of a jet-precooled compressor blade as described in claim 1, characterized in that: The specific construction steps of the digital twin method are as follows. Based on operating condition data and damage data, comparative and aggregated assessments are conducted to obtain a mapping baseline and a digital profile of the object. Based on the mapping baseline and the digital profile of the object, bidirectional synchronization and deviation correction are performed through self-consistent calibration to obtain the online simulation channel and the calibrated state set. The bidirectional synchronization and deviation correction refers to calibrating the online simulation channel and observation set based on the mapping baseline and the digital profile of the object, correcting the deviation sources item by item according to the self-consistent calibration, and updating the online simulation channel and the calibrated state set by using the cooling parameter sequence of the jet pre-cooling intervention and the measured temperature and strain information after cooling. Based on the online simulation channel and the calibrated state set, a closed-loop operation is performed on health assessment, predictive simulation and intervention simulation to obtain driving instructions, early warning list and interpretable key points, and generate a digital twin method.

3. The method for predicting the lifespan of a jet-precooled compressor blade as described in claim 2, characterized in that: Based on remaining lifetime and damage status data, a digital twin method is used to weigh the failure risk and efficiency loss of the blade's health status online. A conservative exploration strategy is then employed to generate jet pre-cooling intervention commands. The specific steps are as follows: Based on remaining life and damage status data, a comprehensive assessment of the current health status of the compressor's last-stage blades is conducted to obtain a quantitative relationship between the blade failure risk level and the degree of operational efficiency loss. Based on the quantitative relationship, a conservative exploration strategy is used to balance and optimize the cooling intensity and injection timing of the jet precooling, and a jet precooling intervention command is generated.

4. The method for predicting the lifespan of a jet-precooled compressor blade as described in claim 3, characterized in that: The specific steps for executing the jet pre-cooling intervention command and dynamically adjusting the cooling parameters to obtain the measured temperature and strain information after cooling collected by the sensor are as follows. Based on the jet pre-cooling intervention command, the jet opening, pulse frequency and nozzle selection are issued and synchronized to form an initial cooling response signal; By using the initial cooling response signal, the mass flow rate, supply air temperature and duty cycle are adjusted in a closed loop using self-consistent calibration to obtain the cooling parameter sequence. Based on the cooling parameter sequence, time-series sampling and integrity verification are performed to obtain the measured temperature and strain information after cooling.

5. The method for predicting the lifespan of a jet-precooled compressor blade as described in claim 4, characterized in that: The cooling parameter sequence refers to a set of adjustable controllable quantities used to control the intensity and distribution of the jet precooling effect, including mass flow rate, supply air temperature, injection pressure and intensity, nozzle selection, injection timing, pulse frequency and duty cycle.

6. The method for predicting the lifespan of a jet-precooled compressor blade as described in claim 5, characterized in that: The blade material parameters are corrected based on measured temperature and strain information, and the corrected blade material parameters are then updated to obtain the updated life prediction results. The specific steps are as follows: Based on measured temperature and strain information, a uniformity method is used to clean, time-align and identify steady-state sections of multi-source measurement data to obtain baseline sequences and load condition data. Based on baseline sequence and load condition data, self-consistent calibration is used to identify and correct deviations in elastic, strength and creep material parameters to obtain a corrected set of blade material parameters. Based on the revised blade material parameter set, cyclic damage and high-temperature degradation are synergistically optimized and continuously corrected to obtain updated life prediction results.

7. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the jet precooling compressor blade life prediction method according to any one of claims 1 to 6.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the jet precooling compressor blade life prediction method according to any one of claims 1 to 6.

Images