A multi-stage rotating cascade particle deposition method and system

By using a multi-stage rotating blade cascade particle deposition stacking experimental method, the problem of being unable to separate inter-stage influences and quantitatively assess turbine blade life in existing technologies has been solved. This method enables the simultaneous positioning of deposition and erosion zones on the same blade, providing accurate blade service life assessment data.

CN122709121BActive Publication Date: 2026-11-17CIVIL AVIATION UNIV OF CHINA
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202611199680.3
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2026-08-10
Publication Date
2026-11-17
Estimated Expiration
2046-08-10

AI Technical Summary

Technical Problem

Existing experimental setups cannot separate interstage effects in multi-stage blade environments, cannot distinguish between deposition zones and erosion zones on the same blade, and cannot quantitatively assess the service life of turbine blades, resulting in discrepancies between assessment results and actual physical processes.

Method used

A multi-stage rotating blade cascade particle deposition stacking experiment method was adopted. The blade profile was scanned by white light interferometer, and combined with three-dimensional scanning and differential calculation, the independent deposition amount of each stage of the blade and the mutual influence between stages were separated. The deposition erosion zone was observed by scanning electron microscopy, the sediment detachment data was corrected, and a quantitative assessment of the blade service life was generated.

Benefits of technology

It enables the simultaneous positioning of the deposition zone, erosion zone, and critical transition zone on the same blade, providing experimental verification data on the depositional evolution law of multi-stage blade cascades and ensuring the accuracy of turbine blade service life assessment.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122709121B_ABST
    Figure CN122709121B_ABST
Patent Text Reader

Abstract

This application relates to the field of aero-engine and power machinery technology, and discloses a method and system for experimental stacking of multi-stage rotating blade cascade particle deposition. The method includes: scanning the initial three-dimensional profile of the blade, measuring the initial mass and particle parameters to form experimental baseline data; constructing a single-stage rotating blade cascade based on the baseline data to obtain the single-stage deposition thickness distribution; extending to two-stage and fully multi-stage configurations, obtaining the deposition thickness distribution for each stage; performing layer-by-layer difference analysis on the data of each stage to separate independent deposition amounts; performing microscopic morphology and composition analysis on key areas to generate zonal distribution data; calculating the mass closure error, correcting the data of the detachment zone, and combining the independent deposition amounts to generate a quantitative assessment result of the blade's service life. This application effectively improves the accuracy of quantitative assessment of blade service life.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This application relates to the field of aero-engine and power machinery technology, and in particular to an experimental method and system for the step-by-step superposition of multi-stage rotating blade cascade particle deposition. Background Technology

[0002] When aero-engines operate in dusty environments, solid particles in the air are carried by high-speed airflow into the compressor and turbine components, where they undergo complex physical interactions with the rotating blade walls. This leads to the removal or adhesion of material from the blade surface, altering the blade profile, clogging film cooling vents, and affecting engine performance and service life. To study blade surface deposition phenomena under rotating conditions, existing blade deposition experimental setups typically include a high-pressure gas source, particle supply device, spray gun, specimen fixture, drive motor, and enclosed chamber, enabling basic functions such as particle acceleration and injection, deposition, and loss measurement.

[0003] In a multi-stage blade cascade environment, particles must be transported sequentially through each stage of the blades. The preceding blades not only capture particles themselves, but their wakes and secondary flows alter the incoming flow conditions faced by subsequent blades, leading to fundamental differences in deposition patterns between the preceding and subsequent stages. However, existing experimental setups employ a single-stage, static, isolated blade cascade configuration, physically severing the information transmission chain of the interstage flow field and completely excluding the interference effect of preceding stages on subsequent stages' deposition. This results in a lack of direct experimental data to support the axial evolution of depositional distribution in multi-stage blade cascades. Regarding measurement methods, current techniques rely on weighing the entire blade before and after the experiment. This overall weighing method only outputs the algebraic sum of depositional weight gain and erosion weight loss, failing to distinguish between the often coexisting depositional adhesion zone, erosion removal zone, and the critical dynamic equilibrium zone on the same blade surface. The spatial location, area proportion, and boundary evolution of these zones are crucial information determining the blade profile degradation path.

[0004] The overlapping of interstage information truncation at the configuration level and spatial distribution averaging at the measurement level makes it impossible for existing technologies to quantitatively separate the independent depositional contributions and interstage interactions of each blade stage from the overall distribution through experimental methods. Furthermore, it is impossible to distinguish between depositional and erosion zones on the same blade and locate their dynamic equilibrium critical boundaries, and it is also impossible to identify and correct for sediment detachment during experiments. These deficiencies in experimental capabilities directly result in a lack of reliable data sources for quantitative assessment of turbine blade service life, leading to discrepancies between assessment results and actual physical processes, and compromising accuracy. Therefore, there is an urgent need for an experimental method and system capable of progressively separating interstage effects, simultaneously characterizing the full-surface deposition and erosion distribution of a single blade, and correcting for detachment data, in order to establish a complete data foundation from multi-stage blade deposition experiments to quantitative assessment of blade service life. Summary of the Invention

[0005] This application provides a multi-stage rotating blade cascade particle deposition stacking experimental method and system to improve the accuracy of quantitative assessment of blade service life.

[0006] In a first aspect, this application provides a multi-stage rotating cascade particle deposition stacking experimental method, including: S1. The initial three-dimensional profile of the turbine experimental blade was scanned using a white light interferometer to establish a profile registration dataset. The initial mass of the turbine experimental blade and the particle size and density parameters were measured and integrated to form the turbine blade deposition experiment reference data. S2. Based on the benchmark data of turbine blade deposition experiments, a single-stage rotating blade configuration of turbine was constructed. The operating speed of the drive motor and the particle ejection outlet velocity were set. The blades were scanned, registered and differentially calculated to obtain the deposition thickness distribution data of the single-stage turbine blade. S3. Add a stationary blade disk to the single-stage rotating blade cascade configuration of the turbine to form a two-stage rotating blade cascade configuration of the turbine. Scan the registration difference and perform deposition falloff correction to obtain the deposition thickness distribution data of the two-stage turbine blades. S4. Assemble all bladed disks to form a full multi-stage rotating bladed cascade configuration of the turbine, scan the three-dimensional profile of each stage blade and perform deposition shedding correction to obtain the deposition thickness distribution data of the full multi-stage turbine blades. S5. Perform layer-by-layer differential calculations on the deposition thickness distribution data of single-stage, two-stage, and multi-stage turbine blades to separate the independent deposition amount of the second-stage blades and the deposition amount of the third-stage and above blades, and generate a blade differential distribution set. S6. Based on the differential distribution set of the blade, the preliminary region is divided, and the morphology of the preliminary region is observed by scanning electron microscopy and the energy spectrum composition is analyzed to generate the distribution data of the blade deposition and erosion zones. S7. Based on the distribution data of blade depositional erosion zones, calculate the mass closure error between the three-dimensional scanning mass change and the measured mass change by the balance. Based on the closure error, extract the detachment zone data in the distribution data of blade depositional erosion zones. Use the neighborhood equivalence method to correct the detachment zone data and verify its accuracy. Combine the independent deposition amount of the second-stage blade and the deposition amount of the third-stage and above blades to generate a quantitative assessment of the service life of the turbine blade.

[0007] Secondly, this application provides a multi-stage rotating cascade particle deposition stacking experimental system, comprising: The benchmark database module is used to scan the initial three-dimensional profile of the turbine experimental blade using a white light interferometer, establish a profile registration dataset, determine the initial mass of the turbine experimental blade and the particle size and density parameters, and integrate them to form benchmark data for turbine blade deposition experiments. The single-stage deposition module is used to build a single-stage rotating blade configuration of a turbine based on the benchmark data of turbine blade deposition experiments, set the operating speed of the drive motor and the particle ejection outlet velocity, scan the blade and register and perform differential calculations to obtain the deposition thickness distribution data of the single-stage turbine blade. The two-stage deposition module is used to add a stator disk to the single-stage rotating blade cascade configuration of the turbine to form a two-stage rotating blade cascade configuration of the turbine. It scans and registers the difference and performs deposition loss correction to obtain the deposition thickness distribution data of the two-stage turbine blades. The full-stage deposition module is used to assemble all bladed disks into the turbine's two-stage rotating blade cascade configuration to form a turbine's full multi-stage rotating blade cascade configuration. It scans the three-dimensional profile of each stage of blades and performs deposition shedding correction to obtain the deposition thickness distribution data of the full multi-stage turbine blades. The deposition separation module is used to perform layer-by-layer differential calculations on the deposition thickness distribution data of single-stage, two-stage, and multi-stage turbine blades, separating the independent deposition amount of the second-stage blades and the deposition amount of the third-stage and above blades, and generating a blade differential distribution set. The partitioning analysis module is used to divide the preliminary regions based on the differential distribution set of the blades, perform scanning electron microscopy morphology observation and energy dispersive spectroscopy analysis on the preliminary regions, and generate leaf deposition and erosion partitioning distribution data. The service life assessment module is used to calculate the mass closure error between the three-dimensional scanning mass change and the actual measured mass change based on the blade deposition erosion zoning distribution data. Based on the closure error, the module extracts the detachment zone data from the blade deposition erosion zoning distribution data, corrects the detachment zone data using the neighborhood equivalence method and verifies its accuracy, and generates a quantitative assessment of the turbine blade service life by combining the independent deposition amount of the second-stage blade and the deposition amount of the third-stage and above blades.

[0008] Thirdly, this application provides a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the above-described multi-stage rotating cascade particle deposition stacking experimental method.

[0009] Compared with the prior art, the beneficial effects of this application are at least as follows: 1. By superimposing configurations and performing differential calculations, the independent depositional amounts of each blade level and the mutual influence between levels are quantitatively separated from the overall distribution, providing experimental verification data for the depositional evolution of multi-level blades.

[0010] 2. Through three-dimensional profile difference and multi-dimensional progressive analysis, the deposition zone, erosion zone and critical transition zone are simultaneously located on the same blade to obtain the spatial distribution of thickness variation across the entire surface, supporting the analysis of blade profile degradation path.

[0011] 3. By calculating the mass closure error and performing equivalent correction in the neighborhood, the three-dimensional scanning data and the actual mass measured by the balance are mutually verified to correct the underestimation caused by sediment detachment and ensure the reliability of the total sediment assessment and lifetime prediction results.

[0012] 4. Through step-by-step experiments and data fusion from single-stage to multi-stage, a complete calibration system from particle impact criteria to cascade deposition distribution is established, enabling the evaluation of turbine blade service life to be transformed from empirical assumptions to quantitative calculations constrained by experimental data. Attached Figure Description

[0013] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0014] Figure 1 This is a flowchart of an experimental method for the step-by-step stacking of multi-stage rotating cascade particle deposition according to this application; Figure 2 This is a schematic diagram of the structure of a multi-stage rotating cascade particle deposition stacking experimental system according to this application. Detailed Implementation

[0015] This application provides a method and system for the step-by-step stacking of multi-stage rotating cascade particle deposition. The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein. Furthermore, the terms "comprising" or "having" and any variations thereof are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0016] For ease of understanding, the specific process of the embodiments of this application is described below. Please refer to [link / reference]. Figure 1 One embodiment of the multi-stage rotating cascade particle deposition stacking experimental method in this application includes: Step S1: Use a white light interferometer to scan the initial three-dimensional profile of the turbine experimental blade, establish a profile registration dataset, determine the initial mass of the turbine experimental blade and the particle size and density parameters, and integrate them to form the turbine blade deposition experiment reference data.

[0017] In one specific embodiment, the process of performing step S1 may specifically include the following steps: The turbine blade and particles of the aero-engine to be tested were obtained. The reference pit in the non-working area of ​​the turbine blade was used as the registration reference. The initial three-dimensional profile of the turbine blade was scanned by a white light interferometer to establish a profile registration dataset. The initial mass of the turbine blades was determined using a high-precision electronic balance, and the particle size and density parameters were determined using a laser particle size analyzer and a densitometer, respectively. By integrating the profile registration dataset, the initial mass of the turbine blades, and the particle size and density parameters, a unified benchmark dataset for turbine blade deposition experiments is formed.

[0018] Specifically, the turbine blade and particles from the aero-engine to be tested are acquired, and a reference recess is machined in the non-working area of ​​the turbine blade. This reference recess has a hemispherical geometry with both diameter and depth on the sub-millimeter scale. The three-dimensional coordinates of the center of this reference recess serve as a spatial coordinate anchor point, providing a geometric reference for the global rigid registration of the three-dimensional surface scanning data before and after the subsequent experiment. This unifies the point cloud data obtained from the two scans into the same coordinate system, eliminating systematic errors caused by clamping and resetting, thereby establishing a permanent geometric reference for point-by-point differential measurements of deposition and erosion.

[0019] The initial three-dimensional profile of the turbine blade was scanned using a white-light interferometer. This white-light interferometer, based on the principle of light coherence, obtains surface height information through phase analysis of interference fringes. A multi-scale strategy was employed for scanning. A low-resolution mode with a lateral resolution of 10 μm was used to globally scan the turbine blade, extracting the coordinates of the center of the reference pit and the opening contour to construct reference feature data for field-of-view stitching. Then, a high-resolution mode with a lateral resolution of 1 μm was used to scan the blade's working surface field by field, acquiring local height data within each field of view. The field-of-view data were stitched together with the center of the reference pit as the reference point. An iterative nearest-point algorithm was used for coordinate transformation. This algorithm solves for the optimal rotation matrix and translation vector by minimizing the sum of squared distances between corresponding points in the source and target point clouds. The maximum number of iterations was set to 200, the distance threshold to 0.01 mm, and the convergence condition to be that the error change between two adjacent iterations is less than 0.001 mm. Reserved reference pit field-of-view data not used for stitching was used as test data to verify the registration accuracy. The initial height values ​​of each coordinate point on the blade surface were recorded, denoted as . The point cloud data, the geometric parameters of the reference pit, and the registration transformation matrix are integrated to establish a surface registration dataset.

[0020] The initial mass of the turbine blade was determined using a high-precision electronic balance with a resolution of 0.01 mg. Multiple independent weighings were performed on the turbine blade under constant temperature and humidity conditions, and the weighing values ​​were recorded. Grubbs' criterion was used to detect outliers in the weighing value sequence. With a significance level of 0.05, a statistical statistic was calculated and compared with a critical value. After removing outliers caused by airflow disturbances or electrostatic adsorption, the arithmetic mean of the remaining weighing values ​​was taken to obtain the initial mass, denoted as . This initial mass serves as a benchmark for subsequent mass closure verification, and is used to compare it with the post-experimental mass and the mass converted from the 3D scan volume.

[0021] The particle size was determined using a laser particle size analyzer. Based on Mie scattering theory, this analyzer emits a 633nm laser beam to irradiate a particle swarm, and a detector array collects light intensity distribution signals at different scattering angles. A non-negative least squares inversion algorithm was used to process the light intensity distribution data. The inversion iteration count was set to 1000, and the regularization parameter was 0.01. The volume equivalent diameter distribution of the particle swarm was solved, and statistical characteristic values ​​of the particle size distribution, including D10, D50, and D90, were extracted. D50 was used as the representative particle size parameter, denoted as D10. The density parameters of particles are determined using a hydrometer. This hydrometer is based on the gas displacement method or the specific gravity bottle method, measuring the volume of liquid displaced or the volume of gas expanded by particles in a container of known volume. The true density is calculated by the ratio of mass to volume. Multiple repeated measurements are performed, and the statistical average is taken to obtain the density parameters.

[0022] A baseline dataset for turbine blade deposition experiments was created by integrating the profile registration dataset, initial mass, grain size, and density parameters. This integration process used the blade number as the primary index key and the blade surface spatial coordinates (x, y) as the secondary index key to align and correlate the height, initial mass, grain size, and density data from the profile registration dataset. Physical parameters with different dimensions were normalized: height data was normalized to the order of the baseline pit depth, mass data to the initial mass baseline, and grain size and density data to their respective means, ensuring all parameters were on the same numerical standard. A baseline dataset for turbine blade deposition experiments was established, comprising geometric, mass, and physical property subsets. This dataset provides a zero-baseline reference for subsequent differential calculations, interstage influence separation, and mass closure verification in single-stage, two-stage, and full multi-stage configuration deposition experiments.

[0023] Step S2: Based on the turbine blade deposition experiment benchmark data, construct a turbine single-stage rotating blade configuration, set the driving motor operating speed and particle injection outlet velocity, scan the blade and register and perform differential calculations to obtain the single-stage turbine blade deposition thickness distribution data.

[0024] In one specific embodiment, the process of performing step S2 may specifically include the following steps: Based on the benchmark data of turbine blade deposition experiments, a first-stage moving blade disk of the corresponding specification was matched, and a single-stage rotating blade cascade configuration of a turbine with only the first-stage moving blade disk installed was built on the turbine blade cascade test bench. Based on the single-stage rotating blade configuration of the turbine, the driving motor speed and particle ejection outlet velocity were set according to the turbine blade deposition experiment benchmark data. The inlet temperature and pressure of the experimental bench were kept at preset constant values, and the single-stage blade particle deposition experiment of preset duration was started. After the experiment, the first-stage moving blade disk was disassembled and loose particles were removed by non-contact purging. The complete three-dimensional profile of each blade was scanned using a white light interferometer. Registration and differential calculations were performed using the profile registration dataset to obtain the deposition thickness distribution data of the single-stage turbine blade.

[0025] Specifically, a first-stage moving blade disk of corresponding specifications was matched based on the turbine blade deposition experiment benchmark data. This matching process used the blade geometric parameters recorded in the turbine blade deposition experiment benchmark data, including airfoil chord length, blade height, installation angle, and tenon specifications, as search criteria. First-stage moving blade disks with geometric compatibility and matching aerodynamic design points were selected from the blade disk inventory. This selection ensured the mechanical connection and aerodynamic flow channel matching between the experimental blade and the blade disk. The selected first-stage moving blade disk was installed on the drive shaft of the turbine blade cascade experimental rig, forming a single-stage rotating blade cascade configuration containing only the first-stage moving blade disk through extension shafts and bearing support. This single-stage rotating blade cascade configuration serves as the foundation for subsequent stage-by-stage stacking experiments. Its rig foundation and connection structure remain unchanged in the subsequent two-stage and multi-stage configuration construction to ensure the consistency of the rig geometric benchmark during configuration switching.

[0026] Based on the single-stage rotating blade configuration of the turbine, and according to the particle properties and target collision kinetic energy requirements recorded in the turbine blade deposition experiment baseline data, the driving motor operating speed and particle ejection exit velocity were set. The driving motor operating speed is denoted as n, which determines the tangential velocity of the blade surface; the particle ejection exit velocity is jointly determined by the high-pressure gas source pressure and the Laval nozzle throat area. The inlet temperature and pressure of the experimental platform were kept at preset constant values, which were predetermined by the experimental condition design document and remained unchanged throughout the experiment to eliminate the interference of environmental fluctuations on particle trajectory and deposition behavior. A single-stage blade particle deposition experiment of a preset duration was started. This preset duration was determined by back-calculation based on the target deposition thickness: based on the particle feed rate, particle density, and the area of ​​the blade to be measured, the experimental duration required to achieve the optimal measurement range of the white light interferometer was estimated. Before the experiment, a sample piece of known area was placed in the experimental section. After the estimated duration, the deposition thickness of the sample piece was measured using a white light interferometer to confirm that the thickness fell within an accurately measurable range. If the thickness did not meet the requirement, the feed rate was adjusted or the duration was extended until the target was achieved. During the experiment, the electronic control integrated control unit collected and recorded parameters such as air source pressure, powder feeding rate, drive motor speed and particle injection outlet speed in real time, and adopted multi-parameter closed-loop control to ensure the repeatability of the working conditions.

[0027] After the experiment, the first-stage moving blade disk was disassembled, and loose particles on the surface were removed using a non-contact purging method. This non-contact purging method used a nitrogen jet to purge the blade surface at a set angle. The nitrogen jet angle was... The nozzle was positioned 50 mm from the blade surface, and the purging pressure was controlled at 0.1 MPa ± 0.02 MPa. The purging time did not exceed 10 seconds to avoid mechanical contact that could cause the deposit layer to peel off or the surface to be scratched, thus preserving the adhered deposits and erosion morphology. A white light interferometer was used to scan the complete three-dimensional profile of each blade. The scanning process used the coordinates of the reference pit center recorded in the profile registration dataset as the initial reference. The same multi-scale scanning strategy was applied to the blades after the experiment as before, obtaining the height values ​​of each coordinate point on the blade surface after the experiment, denoted as [missing information]. The dimension of this height value is length.

[0028] Registration and difference calculations were performed using a surface registration dataset. The registration and difference calculations used a reference pit as the registration reference point. Global rigid registration was performed on the 3D point cloud data obtained from two scans before and after the experiment. The iterative nearest-point algorithm was used to solve the coordinate transformation matrix, with a maximum of 200 iterations, a distance threshold of 0.01 mm, and a convergence condition of the error change between two adjacent iterations being less than 0.001 mm. After registration, point-by-point differences were calculated to obtain the thickness change at each location on the blade surface, expressed as: ,in, The initial height values ​​recorded in the registration dataset for the surface are used. The dimension of is length, with positive values ​​corresponding to deposition thickness and negative values ​​corresponding to erosion depth. The thickness variation is mapped to non-uniform grid cells on the blade surface. The thickness variation within each non-uniform grid cell is averaged using an area integral to obtain the measured deposition thickness for each grid cell, expressed as: ,in, For non-uniform mesh elements, the subscripts i and j are the index numbers of the mesh element. This is the area of ​​the unit, with dimensions of length squared. The dimension of the measurement is length. The measured deposition thickness of each grid cell is correlated with the corresponding spatial coordinates to establish single-stage turbine blade deposition thickness distribution data indexed by the flow direction coordinate x and radial coordinate y on the blade surface.

[0029] Step S3: Add a stationary blade disk to the single-stage rotating blade cascade configuration of the turbine to form a two-stage rotating blade cascade configuration of the turbine. Scan the registration difference and perform deposition shedding correction to obtain the deposition thickness distribution data of the two-stage turbine blades.

[0030] In one specific embodiment, the process of performing step S3 may specifically include the following steps: Keeping the test bench foundation and connection structure of the single-stage rotating blade cascade configuration of the turbine unchanged, a stator disk of the corresponding specification is added to form a two-stage rotating blade cascade configuration of the turbine. Based on the two-stage rotating blade configuration of the turbine, the two-stage blade particle deposition experiment was started with the same drive motor operating speed, particle injection outlet velocity, inlet temperature, inlet pressure and experimental duration as the single-stage experiment. After the experiment, the two-stage bladed disks were disassembled in sequence and loose particles were removed by non-contact purging. The complete three-dimensional profile of each stage of the blades was scanned using a white light interferometer. The profile registration dataset was combined with registration difference and deposition shedding correction to obtain the deposition thickness distribution data of the two-stage turbine blades.

[0031] Specifically, while maintaining the test bench foundation and connecting structure of the single-stage rotating blade cascade configuration of the turbine, a corresponding stationary blade disk is added to the extended section of the drive shaft of the test bench foundation to form a two-stage rotating blade cascade configuration of the turbine. This addition process uses the stationary blade disk geometric parameters recorded in the turbine blade deposition experiment reference data as search criteria. Stationary blade disks that are aerodynamically matched and mechanically compatible with the first-stage moving blade disk are selected from the blade disk inventory. The selected stationary blade disks are then installed at a predetermined axial position downstream of the first-stage moving blade disk using bearing supports, forming a coaxial series structure between the first-stage moving blade disk and the stationary blade disk. The axial clearance between the two stages is predetermined by the design documents to ensure the continuity of the interstage flow channel and avoid rotational interference.

[0032] Based on a two-stage rotating blade configuration, the particle deposition experiment was initiated with the same drive motor operating speed, particle injection outlet velocity, inlet air temperature, inlet air pressure, and experimental duration as the single-stage experiment. This initiation process was performed by a multi-parameter closed-loop control unit (ECU). The ECU synchronously read the setpoints of the operating parameters recorded in the single-stage experiment via the control bus and wrote the drive motor operating speed, air source pressure, powder feeding rate, and experimental duration into the control registers of both stages, ensuring complete consistency between the operating parameters of the two stages and the single-stage experiment. During the experiment, an integrated pressure sensor collected the air source pressure at a sampling frequency, a flow sensor collected the powder feeding rate at a sampling frequency, and a speed sensor collected the drive motor operating speed at a sampling frequency. The ECU used a proportional-integral-derivative (PID) algorithm to adjust each parameter in real time, with the proportional coefficient set to 0.5, the integral coefficient set to 0.1, and the derivative coefficient set to 0.01, and the sampling period set to 10ms. This ensured that the fluctuation range of each parameter was controlled within 5% of the set value, guaranteeing the repeatability of the operating conditions between the two independent experiments.

[0033] After the experiment, the two-stage bladed disks were disassembled sequentially, and loose particles were removed using a non-contact purging method. This non-contact purging method was performed separately on the first-stage moving bladed disk and the stationary bladed disk. Nitrogen jets were used to purge the surface of each stage of the blades at a set angle of 45°±5°, with the nozzle 50mm away from the blade surface. The purging pressure was controlled at [insert pressure here]. A single purging session should not exceed 10 seconds to avoid mechanical contact that could cause the deposit layer to peel off or the surface to be scratched, thus preserving the adhered deposits and erosion morphology. A white light interferometer was used to scan the complete three-dimensional profile of each stage of the blades. The scanning process was performed separately on the first-stage moving and stationary blade disks. Using the coordinates of the reference pit center recorded in the profile registration dataset as the initial reference, multi-scale scanning was performed on each stage of the blades after the experiment to obtain the height values ​​of each coordinate point on the surface of each stage of the blades.

[0034] Registration difference and deposition loss correction were performed using the surface registration dataset. This registration difference was performed separately on the first-stage moving and stationary bladed disks. Using the reference pit as the registration reference point, global rigid registration was performed on the 3D point cloud data obtained from two scans before and after the experiment. The coordinate transformation matrix was solved using the iterative nearest-point algorithm, with a maximum of 200 iterations, a distance threshold of 0.01 mm, and a convergence condition of the error change between two adjacent iterations being less than 0.001 mm. After registration, point-by-point subtraction was performed to obtain the thickness change of each blade stage after the experiment. These thickness changes were mapped to non-uniform grid cells on the surface of each blade stage. The thickness change within each non-uniform grid cell was averaged using area integration to obtain the preliminary measured deposition thickness value for each grid cell.

[0035] Deposition detachment correction is performed. This correction identifies detachment areas based on a joint criterion of thickness variation distribution and scanning electron microscopy (SEM) morphology. Specifically, within a non-uniform grid cell, if post-experiment SEM observation reveals cracks, warping, or flaky detachment pits in the sediment layer, and the thickness variation in this area shows a sudden drop or abrupt negative change compared to the surrounding continuous sedimentation area, then the grid cell is marked as a potential detachment area. For grid cells marked as potential detachment areas, the neighborhood equivalence method is used for estimation. This method uses neighboring non-detachment grid cells with the same blade surface properties and under the same aerodynamic impact conditions as the equivalent neighborhood. Before calculating the equivalent thickness, the measured deposition thickness values ​​of each equivalent neighborhood grid cell are normalized by dividing the absolute thickness value of each neighborhood cell by the global maximum deposition thickness of the blade stage, thus unifying the dimensions of the neighborhood data into dimensionless relative values. When setting the weighting coefficients, the neighboring units closer to the potential detachment zone are given higher weights. A distance-inverse weighting strategy is adopted, that is, the weight is inversely proportional to the spatial distance from the center of the neighboring unit to the center of the potential detachment zone. The normalized thickness values ​​of each neighboring unit are weighted and averaged to obtain the equivalent thickness estimate. The equivalent thickness estimate is used as the original deposition thickness estimate of the potential detachment zone to update the global deposition thickness distribution. The corrected deposition thickness distribution of each stage of the turbine blade is associated with the corresponding spatial coordinates to establish two-stage turbine blade deposition thickness distribution data.

[0036] Step S4: Assemble all bladed disks into a turbine full multi-stage rotating bladed cascade configuration by assembling the two-stage rotating blade cascade configuration. Scan the three-dimensional profile of each stage blade and perform deposition shedding correction to obtain the deposition thickness distribution data of the full multi-stage turbine blades.

[0037] In one specific embodiment, the process of performing step S4 may specifically include the following steps: Keeping the test bench foundation and connecting structure of the two-stage rotating blade cascade configuration of the turbine unchanged, the remaining moving blade disks and stationary blade disks of each stage are assembled in sequence to form a full multi-stage rotating blade cascade configuration of the turbine. Based on the turbine multi-stage rotating blade configuration, the drive motor operating speed, particle injection outlet velocity, inlet temperature, inlet pressure and experimental duration were kept the same as those of the single-stage and two-stage blade particle deposition experiments, and the multi-stage blade particle deposition experiment was started. After the experiment, all bladed disks at each stage were disassembled in sequence and loose particles were removed by non-contact purging. The complete three-dimensional profile of each stage of the blades was scanned using a white light interferometer. The profile registration dataset was combined with registration difference and deposition shedding correction to obtain the deposition thickness distribution data of the entire multi-stage turbine blades.

[0038] Specifically, keeping the test bench foundation and connection structure of the two-stage rotating blade cascade configuration of the turbine unchanged, the remaining moving and stationary blade disks of each stage are assembled sequentially. This sequential assembly process uses the geometric parameters of the full multi-stage blade disks recorded in the turbine blade deposition experiment reference data as search criteria. Moving and stationary blade disks of each stage that are aerodynamically matched and mechanically compatible with the upstream installed blade disks are selected from the blade disk inventory. The selected blade disks are then installed one by one onto the drive shaft and the inner wall of the casing according to the axial sequence and clearance requirements predetermined in the design documents, using bearing supports and a casing fixing structure. This creates a multi-stage structure where the rotating and stationary components of each stage are coaxially connected, thus forming the turbine's full multi-stage rotating blade cascade configuration. In this configuration, each stage of the moving blade disks, as rotating components, is driven synchronously by the same drive motor via an extended shaft. Each stage of the stationary blade disks, as stationary components, is fixed to the inner wall of the annular casing. The axial clearance between adjacent stages is predetermined in the design documents to ensure the continuity of the multi-stage flow channels and avoid rotational interference.

[0039] Based on the turbine multi-stage rotating blade configuration, the multi-stage blade particle deposition experiment was started with the same drive motor operating speed, particle ejection outlet velocity, inlet air temperature, inlet air pressure, and experimental duration as the single-stage and two-stage blade particle deposition experiments. This start-up process was executed by a multi-parameter closed-loop control unit. Specifically, the integrated electronic control unit synchronously read the setpoint values ​​of the operating parameters recorded in the single-stage and two-stage experiments via the control bus, and wrote the drive motor operating speed, air source pressure, powder feed rate, inlet air temperature, inlet air pressure, and experimental duration into the control register of the multi-stage experiment with the same values, ensuring that the operating parameters of the multi-stage experiment were completely consistent with the previous two experiments. During the experiment, the integrated pressure sensor collected the air source pressure at a sampling frequency, the flow sensor collected the powder feeding rate at a sampling frequency, the temperature sensor collected the inlet air temperature at a sampling frequency, and the speed sensor collected the driving motor speed at a sampling frequency. The integrated electronic control unit used a proportional-integral-derivative algorithm to adjust the parameters in real time. The proportional coefficient was set to 0.5, the integral coefficient was set to 0.1, the derivative coefficient was set to 0.01, and the sampling period was set to 10ms, so that the fluctuation range of each parameter was controlled within 5% of the set value, ensuring the repeatability of the working conditions in the three independent experiments.

[0040] After the experiment, all stages of the bladed disks were disassembled sequentially, and loose particles were removed using a non-contact purging method. This non-contact purging method was performed on each stage of the bladed disks individually, using a nitrogen jet at a set angle of 45°±5°, with the nozzle 50mm from the blade surface, and the purging pressure controlled at [insert pressure here]. A single purging session should not exceed 10 seconds to avoid mechanical contact that could cause the deposit layer to peel off. A white light interferometer was used to scan the complete three-dimensional profile of each blade stage. The scanning process was performed separately on all blade stages. Using the coordinates of the center of the reference pit of each blade stage recorded in the profile registration dataset as the initial reference, multi-scale scanning was performed on each blade stage after the experiment to obtain the height values ​​of each coordinate point on the surface of each blade stage after the experiment.

[0041] Registration difference and deposition loss correction were performed using the surface registration dataset. This registration difference was performed separately for all bladed disks of each grade. Using the reference pits on each blade as registration reference points, global rigid registration was performed on the 3D point cloud data obtained from two scans before and after the experiment. Specifically, the iterative nearest-point algorithm was used to solve the coordinate transformation matrix, with a maximum of 200 iterations, a distance threshold of 0.01 mm, and a convergence condition of the error change between two adjacent iterations being less than 0.001 mm. After registration, point-by-point subtraction was performed to obtain the thickness change of each blade after the experiment. This thickness change was mapped to non-uniform grid cells on the surface of each blade. The thickness change within each non-uniform grid cell was averaged using area integration to obtain the preliminary measured deposition thickness value for each grid cell.

[0042] Deposition detachment correction is performed. This correction identifies detachment areas based on a joint criterion of thickness variation distribution and scanning electron microscopy (SEM) morphology. Specifically, within a non-uniform grid cell, if post-experiment SEM observation reveals cracks, warping, or flaky detachment pits in the sediment layer, and the thickness variation in this area shows a sudden drop or abrupt negative change compared to the surrounding continuous sedimentary areas, then the grid cell is marked as a potential detachment area. For grid cells marked as potential detachment areas, the neighborhood equivalence method is used for estimation. Specifically, neighboring non-detachment grid cells with the same blade surface properties and under the same aerodynamic impact conditions as the potential detachment area are used as equivalent neighbors. Before calculating the equivalent thickness, the measured sediment thickness values ​​of each equivalent neighbor grid cell are normalized by dividing the absolute thickness value of each neighbor cell by the global maximum sediment thickness of the blade at that stage, thus unifying the dimensions of the neighbor data into dimensionless relative values. When setting the weighting coefficients, the principle of spatial distance attenuation is applied to make the neighboring units closer to the potential detachment zone have higher weights. Specifically, the inverse distance weighting strategy is adopted, that is, the weight is inversely proportional to the spatial distance from the center of the neighboring unit to the center of the potential detachment zone. The distance index is set to 1.0. The normalized thickness values ​​of each neighboring unit are weighted and averaged to obtain the equivalent thickness estimate. The equivalent thickness estimate is used as the original deposition thickness estimate of the potential detachment zone to update the global deposition thickness distribution. The corrected deposition thickness distribution of each stage of the blade is associated with the corresponding spatial coordinates to establish the deposition thickness distribution data of the entire multi-stage turbine blade.

[0043] Step S5: Perform layer-by-layer differential calculations on the deposition thickness distribution data of single-stage, two-stage, and multi-stage turbine blades to separate the independent deposition amount of the second-stage blades and the deposition amount of the third-stage and above blades, and generate a blade differential distribution set.

[0044] In one specific embodiment, the process of performing step S5 may specifically include the following steps: Acquire deposition thickness distribution data for single-stage turbine blades, two-stage turbine blades, and all multi-stage turbine blades, and establish a spatial data matrix of deposition thickness for all multi-stage turbine blades. Based on the spatial data matrix of full multi-stage deposition thickness, the deposition thickness values ​​of the corresponding spatial positions of turbine blades in adjacent configuration levels are calculated by difference to obtain the reverse influence of the newly added turbine blades on the deposition of existing turbine blades. The deposition thickness distribution of the second-stage turbine blade is directly extracted from the deposition thickness distribution data of the two-stage turbine blade to obtain the independent deposition amount of the second-stage blade; the deposition thickness distribution of the third-stage and above blades is directly extracted from the deposition thickness distribution data of the entire multi-stage turbine blade to obtain the deposition amount of the third-stage and above blades. By integrating the reverse influence amount, the independent deposition amount of the second-level blade, the deposition amount of the third-level and above blades, and the corresponding spatial coordinate information, and removing abnormal measurement points and missing data points, a blade differential distribution set containing the spatial coordinates of the blade surface and the corresponding thickness difference value is generated.

[0045] Specifically, deposition thickness distribution data for single-stage turbine blades, two-stage turbine blades, and all multi-stage turbine blades were acquired to establish a spatial data matrix for all multi-stage deposition thickness. This spatial data matrix uses the flow-direction and radial coordinates of the non-uniform grid cells on the blade surface as spatial indices and the configuration level as a hierarchical index. Matrix elements are the measured deposition thickness values ​​within the corresponding grid cells after shedding correction. The measured deposition thickness values ​​of the corresponding grid cells for each stage of the blades in single-stage, two-stage, and all multi-stage configurations are arranged in a structured manner to form a three-dimensional data array with spatial coordinates and configuration level as dimensions. In this data array, the deposition thickness distribution data for single-stage turbine blades corresponds to the grid cell deposition thickness when only the first-stage moving blade disk exists; the deposition thickness distribution data for two-stage turbine blades corresponds to the grid cell deposition thickness of each stage of the blades when the first-stage moving blade disk and the stationary blade disk coexist; and the deposition thickness distribution data for all multi-stage turbine blades corresponds to the grid cell deposition thickness of each stage of the blades when all blade disks coexist. The planar dimensions of the grid cells are set according to the blade curvature and deposition characteristic gradient. Non-uniform division is adopted in the flow direction and radial direction. The thickness accuracy within each grid is guaranteed by a white light interferometer, reaching the submicron level.

[0046] Layer-by-layer differential calculations were performed on the spatial data matrix of the full multi-stage sedimentary thickness. This calculation was performed progressively from low to high configuration levels, algebraically subtracting the measured sedimentary thickness values ​​at the same spatial coordinate position in adjacent configuration levels. The sedimentary thickness distribution of the first-stage moving blade disk in a two-stage configuration was subtracted from that in a single-stage configuration, cell-by-cell, to obtain the reverse influence of the stationary blade disk on the first-stage moving blade disk sedimentation. The sedimentary thickness distribution of the stationary blade disk in the two-stage configuration was directly extracted as the independent sedimentary amount of the second-stage blade. This independent sedimentary amount of the second-stage blade is the sedimentary thickness distribution formed by particles intercepted by the stationary blade disk itself.

[0047] The depositional thickness distribution of the first-stage moving bladed disk in the fully multi-stage configuration is subtracted from the depositional thickness distribution of the first-stage moving bladed disk in the two-stage configuration on a grid-by-grid basis to obtain the reverse influence of subsequent bladed disks on the first-stage moving bladed disk. Similarly, the depositional thickness distribution of the stationary bladed disk in the fully multi-stage configuration is subtracted from the depositional thickness distribution of the stationary bladed disk in the two-stage configuration on a grid-by-grid basis to obtain the reverse influence of subsequent bladed disks on the stationary bladed disk. The depositional thickness distribution of subsequent bladed disks in the fully multi-stage configuration is directly extracted as the depositional amount of the third-stage and above blades. This depositional amount of the third-stage and above blades is the depositional thickness distribution formed by the interception of particles by the subsequent bladed disks themselves.

[0048] This process integrates the independent deposition amounts of second-level blades, third-level and above blades, and their corresponding spatial coordinates. Specifically, it aligns the independent deposition amounts and their inverse effects according to blade number, level number, and coordinates of non-uniform grid cells, establishing a multi-level differential dataset indexed by spatial coordinates. Outlier and missing data points are then removed. Specifically, a statistical outlier detection algorithm is used to clean the multi-level differential dataset. The global mean and standard deviation of the difference values ​​for each grid cell are calculated. Data points deviating from the mean by more than three times the standard deviation are marked as outliers and removed. For missing data points caused by scanning blind spots or registration failures, an inverse distance weighted interpolation algorithm based on neighboring grid cells is used for filling. The interpolation radius is set to three grid cells, the distance exponent is set to 2.0, and the weight is inversely proportional to the spatial distance from the center of the grid cell to the center of the missing point. A blade differential distribution set containing the spatial coordinates of the blade surface and the corresponding thickness difference values ​​is generated. The blade differential distribution set refers to the data set that records the changes in deposition thickness and their spatial location within each grid cell of each blade level, providing input for subsequent determination of depositional erosion zoning and lifetime assessment.

[0049] Step S6: Based on the differential distribution set of the blade, divide the preliminary region, perform scanning electron microscopy morphology observation and energy dispersive spectroscopy analysis on the preliminary region, and generate blade depositional erosion zoning distribution data.

[0050] In one specific embodiment, the process of performing step S6 may specifically include the following steps: Based on the differential distribution set of the blades, and combined with the positive and negative values ​​and amplitude range of the thickness difference at each point on the surface of each blade, the surface of each blade is divided into a preliminary region containing the deposition region, the erosion region and the dynamic equilibrium critical region. Scanning electron microscopy morphology observation and energy dispersive spectroscopy analysis were carried out sequentially on the preliminary area to simultaneously acquire surface micromorphology features and elemental composition information under the same field of view; By combining surface micromorphological features and elemental composition information, the depositional regions, erosion regions, and dynamic equilibrium critical regions on the surface of blades at all levels are identified, and the distribution data of blade depositional erosion zones are integrated to generate data.

[0051] Specifically, based on the blade differential distribution set, and combining the sign and amplitude range of the thickness difference at each point on the blade surface of each level, the blade surface of each level is divided into preliminary regions containing deposition areas, erosion areas, and dynamic equilibrium critical areas. This blade differential distribution set refers to the data set recording the changes in deposition thickness and their spatial coordinates within each non-uniform grid cell of each blade level. The thickness difference refers to the thickness change obtained by subtracting points before and after the experiment using white light interferometry scanning registration. To statistically separate random noise from real deposition or erosion signals, the 3σ criterion is used to set the discrimination threshold. Specifically, the standard deviation of the thickness values ​​obtained by repeated measurements on the same blade surface in a non-working reference area not involved in the experiment using a white light interferometer is used as the benchmark, and three times this standard deviation is used as the discrimination threshold. A preliminary determination is made based on the sign and magnitude of the thickness difference relative to a distinguishing threshold: when the thickness difference is greater than the distinguishing threshold and multiple consecutive points have positive values, the point is determined to belong to a depositional region; when the thickness difference is less than the negative distinguishing threshold and multiple consecutive points have negative values, the point is determined to belong to an erosion region; when the absolute value of the thickness difference is less than or equal to the distinguishing threshold and the thickness differences of adjacent points fluctuate between positive and negative, the point is determined to belong to a dynamic equilibrium critical region. The above point-by-point determination results are mapped to non-uniform grid cells. If the proportion of points determined to be depositional regions in a certain grid cell exceeds 50%, the entire grid cell is classified as a depositional region; if the proportion of points determined to be erosion regions exceeds 50%, the entire grid cell is classified as an erosion region; all other cases are classified as dynamic equilibrium critical regions. This yields preliminary region division results covering the entire surface of each blade.

[0052] Scanning electron microscopy (SEM) morphology observation and energy dispersive spectroscopy (EDS) composition analysis were sequentially performed on the initial area. The SEM is an instrument that uses a focused electron beam to scan the sample surface and acquires the surface microstructure by detecting secondary electron or backscattered electron signals. Its accelerating voltage is set to 15kV to 20kV, the working distance to 10mm to 15mm, and the magnification to be adjusted from 500x to 5000x according to observation requirements. EDS composition analysis is a micro-area composition analysis technique used in conjunction with SEM. It identifies the types and contents of elements within a micro-area by detecting the characteristic X-rays generated after the sample is excited by the electron beam. To simultaneously acquire surface microstructure features and elemental composition information in the same field of view, the specific implementation is as follows: within the same observation field of view of the SEM, the surface morphology image is first acquired in secondary electron mode, then switched to EDS acquisition mode, with a characteristic X-ray count rate of not less than 1000cps as the acquisition condition, and the acquisition time set to 60 to 120 seconds, to acquire the elemental composition information within this field of view. Representative observation points were selected for each grid unit in the preliminary area, and observations were conducted sequentially in the order of first the sedimentary area, then the erosion area, and finally the dynamic equilibrium critical area to avoid cross-contamination.

[0053] By combining surface microstructure characteristics and elemental composition information, deposition areas, erosion areas, and dynamic equilibrium critical areas on the surface of blades at various stages are identified. The surface microstructure characteristics refer to the surface texture and structural features presented in scanning electron microscope (SEM) images, while the elemental composition information refers to the integrated intensity data of characteristic peaks of each element obtained from energy dispersive spectroscopy (EDS) analysis. For deposition areas, the surface microstructure characteristics are characterized by a layer of accumulated particles, with visible individual particle outlines and interparticle adhesion traces; the original processing texture of the substrate is not visible. The elemental composition information shows characteristic elements of particulate materials, and their content is significantly higher than the baseline level of the same element in the substrate. This baseline level was determined by EDS analysis of the same area of ​​undeposited blades before the experiment. For erosion areas, the surface microstructure characteristics are characterized by directional cutting grooves or fish-scale-like plastic deformation pits on the surface; the substrate texture is destroyed and there is no foreign particle adhesion; the groove direction is consistent with the local airflow velocity vector direction. The elemental composition information only shows characteristic elements of the substrate, and their content ratio is consistent with the baseline level before the experiment; no characteristic elements of particulate materials are detected. For the critical region of dynamic equilibrium, the surface micromorphology features the simultaneous presence of particle adhesion and erosion traces. Some areas show event superposition, i.e., erosion pits contain embedded new deposited particles or traces of cutting and peeling remain on the deposited layer. The substrate texture is partially visible and partially covered. The elemental composition information contains characteristic elements of both the substrate and particulate materials. The elemental content ratio is defined as the ratio of the integrated intensity of the characteristic peak of the particulate material's characteristic element to the integrated intensity of the characteristic peak of the substrate's characteristic element. When this ratio is greater than 1, it indicates that the particulate material is dominant in this micro-region; less than 1 indicates that the substrate is dominant; and approximately equal to 1 indicates that the two are of similar magnitude, reflecting the degree of dynamic equilibrium between deposition and erosion. The above identification results are aligned and correlated according to grid cell coordinates and level numbering to generate blade deposition and erosion zoning distribution data. This data refers to the set of data recording the region type and corresponding micromorphological and elemental evidence of each grid cell of each level of blade.

[0054] Step S7: Based on the distribution data of blade depositional erosion zones, calculate the mass closure error between the three-dimensional scanning mass change and the measured mass change of the balance. Based on the closure error, extract the detachment zone data in the distribution data of blade depositional erosion zones. Use the neighborhood equivalence method to correct the detachment zone data and verify its accuracy. Combine the independent deposition amount of the second-stage blade and the deposition amount of the third-stage and above blades to generate a quantitative assessment of the service life of the turbine blade.

[0055] In one specific embodiment, the process of performing step S7 may specifically include the following steps: The measured mass of turbine blades after single-stage, two-stage, and multi-stage blade particle deposition experiments was obtained, and the change in measured mass was calculated in combination with the initial mass of the turbine blades. Based on the distribution data of blade deposition erosion zones and the initial mass of turbine experimental blades, the three-dimensional scanning mass change is calculated by combining the deposition thickness distribution data of single-stage, two-stage, and multi-stage turbine blades, and the mass closure error between the three-dimensional scanning mass change and the measured mass change by the balance is calculated. Based on the mass closure error, the data of the detachment zone in the blade deposition erosion zoning distribution data is extracted. The detachment zone data is corrected by the neighborhood equivalence method. The accuracy of the corrected data is verified by recalculating the mass closure error, and the corrected blade deposition erosion zoning distribution data is obtained. By combining the independent deposition amount of the second-stage blade, the deposition amount of the third-stage and above blades, and the corrected blade deposition erosion zoning distribution data, blade deposition erosion damage parameters are constructed. Based on the blade deposition erosion damage parameters and the preset failure threshold, a quantitative assessment of the service life of the turbine blade is generated.

[0056] Specifically, the measured mass of turbine blades after single-stage, two-stage, and multi-stage blade particle deposition experiments was obtained using a balance. This measured mass refers to the mass data obtained by performing multiple independent weighings of each stage of the turbine blade using a high-precision electronic balance under constant temperature and humidity conditions. The high-precision electronic balance has a resolution of 0.01 mg, and the number of weighings was set to 5 to 7. The Grubbs criterion was used to detect outliers in the multiple weighing values, with a significance level set at 0.05. A statistical measure was calculated and compared with a critical value. After removing outlier values ​​caused by airflow disturbances or electrostatic adsorption, the arithmetic mean of the remaining weighing values ​​was taken to obtain the measured mass. The change in the measured mass was calculated in conjunction with the initial mass of the turbine blade. This change in measured mass is the algebraic difference between the measured mass after the experiment and the initial mass before the experiment. A greater than zero algebraic difference indicates a net mass increase, corresponding to deposition-dominant forces; a less than zero algebraic difference indicates a net mass decrease, corresponding to erosion-dominant forces.

[0057] Based on the distribution data of blade depositional erosion zones and the initial mass of turbine experimental blades, the three-dimensional scanning mass change is calculated by combining the depositional thickness distribution data of single-stage, two-stage, and multi-stage turbine blades. This three-dimensional scanning mass change refers to the estimated mass change obtained by calculating the sediment volume and erosion volume respectively, based on the depositional and erosion zones identified by white light interferometric scanning differential data, and then converting the sediment density to the substrate density. Specifically, for non-uniform grid cells marked as depositional zones in the blade depositional erosion zone distribution data, the measured depositional thickness of each grid cell is multiplied by its area to obtain the depositional volume. The accumulated depositional volumes of each grid cell are then summed to obtain the global depositional volume. This global depositional volume is multiplied by the sediment density to obtain the depositional mass. The same operation is performed on the erosion zones: the erosion depth is multiplied by the area to obtain the erosion volume. The accumulated erosion volumes of each grid cell are then summed to obtain the global erosion volume. This global erosion volume is multiplied by the substrate density to obtain the erosion mass. The three-dimensional scanning mass change is obtained by subtracting the erosion mass from the depositional mass.

[0058] The mass closure error between the 3D scanning mass change and the measured mass change by the balance is calculated. This mass closure error is calculated using the following formula: the absolute value of the difference between the 3D scanning mass change and the measured mass change by the balance is divided by the larger of the two absolute values, and the resulting ratio is expressed as a percentage. The 3D scanning mass change is obtained by multiplying the sediment volume by the sediment density and then subtracting the erosion zone volume by the substrate density. The measured mass change by the balance is obtained by subtracting the initial mass from the post-experimental weighing mass. An allowable error threshold is set; for example, this threshold is set to 10%. When the mass closure error is less than this threshold, the mass estimate of the total sediment erosion obtained from the white light interferometry scan is considered to be in agreement with the measured value by the electronic balance within the error range, and the 3D morphology difference data passes macroscopic mass verification.

[0059] Data on detachment zones in the leaf deposition erosion zoning distribution data is extracted based on mass closure error. This detachment zone data refers to the area formed during the experiment due to localized brittle detachment of the deposited particle layer caused by high-speed airflow impact or continuous particle collision. This detachment leads to an underestimation of the true total deposition amount by changes in 3D scanning mass. Specifically, within a non-uniform grid cell, if scanning electron microscopy reveals cracks, flaking, or flaking pits in the deposition layer, and the thickness variation in this area shows a sudden drop or negative abrupt change compared to the surrounding continuous deposition area, then this grid cell is marked as detachment zone data.

[0060] The neighborhood equivalence method was used to correct the data of the detachment zone. This neighborhood equivalence method refers to the correction method that uses neighboring non-detached grid cells with the same blade surface properties and the same aerodynamic impact conditions as equivalent references, and estimates the original deposition thickness of the detachment zone by weighted averaging. Specifically, the implementation is as follows: Centered on the detachment zone, non-detached grid cells are selected as equivalent neighborhoods within the surrounding area. The neighborhood radius is determined based on the spatial correlation scale of the blade surface depositional characteristics, ensuring coverage of the continuous depositional area surrounding the detachment zone without crossing obvious aerodynamic boundaries. Before weighted averaging, the measured depositional thickness values ​​of each equivalent neighborhood grid cell are normalized by dividing the absolute thickness value of each neighborhood cell by the global maximum depositional thickness of that blade class, unifying the dimensions of each neighborhood data into dimensionless relative values. When setting the weighting coefficient, based on the principle of spatial distance attenuation, neighborhood cells closer to the detachment zone have higher weights. A distance reciprocal weighting strategy is adopted, meaning the weight is inversely proportional to the spatial distance from the center of the neighborhood cell to the center of the detachment zone. The distance index is set to 2.0. The normalized thickness values ​​of each neighborhood cell are then weighted and averaged to obtain an estimated equivalent thickness. This estimated equivalent thickness is used as the original depositional thickness estimate for the detachment zone, updating the global depositional thickness distribution to obtain corrected blade depositional erosion zoning distribution data.

[0061] The accuracy of the corrected data was verified by recalculating the mass closure error. Specifically, the global sediment volume was recalculated based on the corrected sediment thickness distribution, and the corrected 3D scan mass change was obtained by combining it with the sediment density conversion. The mass closure error was then recalculated with the actual mass change measured by the balance. If the corrected mass closure error was lower than that before correction and below the allowable error threshold, it indicated that the correction of the detachment area was reasonable and effective. If it still exceeded the threshold, it suggested the existence of unidentified detachment areas or that the equivalent substitution assumption for the neighboring area was not applicable, and manual verification based on the microstructure was required.

[0062] By combining the independent deposition amounts of the second-stage blades, the deposition amounts of the third-stage and above blades, and the corrected blade depositional erosion zoning distribution data, a blade depositional erosion damage parameter is constructed. This blade depositional erosion damage parameter is a dimensionless index that comprehensively characterizes the degree of depositional and erosion damage on the blade surface. Its calculation formula is: Blade depositional erosion damage parameter = Depositional weight coefficient × (Maximum depositional thickness in critical areas ÷ Maximum allowable depositional thickness in design) + Erosion weight coefficient × (Maximum erosion depth in critical areas ÷ Maximum allowable erosion depth in design). Wherein, the maximum depositional thickness in critical areas is extracted from the independent deposition amounts of the second-stage blades and the deposition amounts of the third-stage and above blades, taking the maximum thickness value in the grid cells of the depositional zone of that stage of blade; the maximum allowable depositional thickness in design is determined by the blade profile allowable thickening limit specified in the blade aerodynamic design documents; the maximum erosion depth in critical areas is extracted from the erosion zone grid cells of the corrected blade depositional erosion zoning distribution data, taking the maximum erosion depth; the maximum allowable erosion depth in design is determined by the wall thickness allowable thinning limit specified in the blade structural strength design documents; the sum of the depositional weight coefficient and the erosion weight coefficient is 1, and their values ​​are determined based on the main controlling factors of blade failure. Specifically, the implementation involves: normalizing the independent deposition amounts of the second-stage blades and the deposition amounts of the third-stage and above blades; dividing the deposition thickness of each stage by the maximum allowable deposition thickness for that stage to obtain a dimensionless deposition damage component; for the erosion areas in the corrected blade deposition erosion zoning distribution data, dividing the erosion depth of each grid cell by the maximum allowable erosion depth for that stage to obtain a dimensionless erosion damage component; and summing the dimensionless deposition damage component and the dimensionless erosion damage component by area weight to obtain the blade deposition erosion damage parameter. A preset failure threshold is set. This preset failure threshold refers to the critical value of the blade deposition erosion damage parameter determined according to the aero-engine turbine blade design specifications and material fatigue limits. When the blade deposition erosion damage parameter reaches or exceeds this preset failure threshold, the blade structural integrity is deemed lost.

[0063] A quantitative assessment of turbine blade service life is generated based on blade deposition erosion damage parameters and a preset failure threshold. This quantitative assessment of turbine blade service life refers to using the cumulative experimental time or equivalent number of cycles corresponding to the blade deposition erosion damage parameters reaching the preset failure threshold as the quantitative result of the service life limit. Specifically, it involves establishing a monotonically increasing relationship between blade deposition erosion damage parameters and cumulative experimental time, and using linear interpolation or least squares fitting to determine the critical time when the damage parameters reach the preset failure threshold. This critical time is the quantitative assessment result of the turbine blade service life.

[0064] It is understood that the executing entity of this application can be a multi-stage rotating cascade particle deposition stacking experimental system, or it can be a terminal or a server; the specific implementation is not limited here. This application's embodiments use a server as an example for illustration.

[0065] The above describes a method for the step-by-step stacking of multi-stage rotating cascade particle deposition in the embodiments of this application. The following describes a system for the step-by-step stacking of multi-stage rotating cascade particle deposition in the embodiments of this application. Please refer to [link to relevant documentation]. Figure 2 One embodiment of the multi-stage rotating cascade particle deposition stacking experimental system in this application includes: The benchmark database module is used to scan the initial three-dimensional profile of the turbine experimental blade using a white light interferometer, establish a profile registration dataset, determine the initial mass of the turbine experimental blade and the particle size and density parameters, and integrate them to form benchmark data for turbine blade deposition experiments. The single-stage deposition module is used to build a single-stage rotating blade configuration of a turbine based on the benchmark data of turbine blade deposition experiments, set the operating speed of the drive motor and the particle ejection outlet velocity, scan the blade and register and perform differential calculations to obtain the deposition thickness distribution data of the single-stage turbine blade. The two-stage deposition module is used to add a stator disk to the single-stage rotating blade cascade configuration of the turbine to form a two-stage rotating blade cascade configuration of the turbine. It scans and registers the difference and performs deposition loss correction to obtain the deposition thickness distribution data of the two-stage turbine blades. The full-stage deposition module is used to assemble all bladed disks into the turbine's two-stage rotating blade cascade configuration to form a turbine's full multi-stage rotating blade cascade configuration. It scans the three-dimensional profile of each stage of blades and performs deposition shedding correction to obtain the deposition thickness distribution data of the full multi-stage turbine blades. The deposition separation module is used to perform layer-by-layer differential calculations on the deposition thickness distribution data of single-stage, two-stage, and multi-stage turbine blades, separating the independent deposition amount of the second-stage blades and the deposition amount of the third-stage and above blades, and generating a blade differential distribution set. The partitioning analysis module is used to divide the preliminary regions based on the differential distribution set of the blades, perform scanning electron microscopy morphology observation and energy dispersive spectroscopy analysis on the preliminary regions, and generate leaf deposition and erosion partitioning distribution data. The service life assessment module is used to calculate the mass closure error between the three-dimensional scanning mass change and the actual measured mass change based on the blade deposition erosion zoning distribution data. Based on the closure error, the module extracts the detachment zone data from the blade deposition erosion zoning distribution data, corrects the detachment zone data using the neighborhood equivalence method and verifies its accuracy, and generates a quantitative assessment of the turbine blade service life by combining the independent deposition amount of the second-stage blade and the deposition amount of the third-stage and above blades.

[0066] Through the collaborative efforts of the aforementioned components, the baseline database module first establishes the initial blade profile and mass baseline, as well as particle parameters, providing a unified registration reference and data zero point for subsequent stage experiments. The single-stage deposition module then completes an independent deposition experiment on the first-stage moving blade disk, obtaining a basic depositional distribution free from interstage interference. The two-stage deposition module inherits the rig foundation and operating condition settings of the single-stage configuration, introducing interstage coupling effects by adding a stationary blade disk to obtain depositional data after the wake of the preceding stage and the secondary flow alters the incoming flow conditions of the subsequent stage. The full-stage deposition module further expands to a complete multi-stage configuration, capturing the entire cascade depositional distribution under the same operating conditions. The depositional separation module performs layer-by-layer differential calculations on the single-stage, two-stage, and full multi-stage data, separating the independent depositional amounts of each stage and the interstage interactions from the total depositional distribution. The zoning analysis module divides the blade surface into gridded regions based on the differential results, and, combined with scanning electron microscopy morphology observation and energy dispersive spectroscopy analysis, simultaneously identifies depositional zones, erosion zones, and dynamic equilibrium critical zones. The life assessment module cross-verifies the mass changes from 3D scanning with the actual mass measured by a balance. It identifies detachment areas through closure error and corrects them using the neighborhood equivalence method. The corrected deposition distribution is then integrated with the independent deposition amounts at each stage to ultimately generate a quantitative assessment of the turbine blade's service life. The modules are sequentially connected, with data passed level by level. The output of the preceding module directly serves as the input for the following module. From baseline establishment to configuration expansion, from distribution separation to region identification, and then to mass verification and detachment correction, a complete workflow is formed from experimental data acquisition to life assessment result output. This enables multi-stage blade deposition experiments to have the ability to quantitatively separate inter-stage influences and identify differences across the entire surface of a single blade.

[0067] This application also provides a computer-readable storage medium, which can be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium, wherein the computer-readable storage medium stores instructions that, when the instructions are executed on a computer, cause the computer to perform the steps of the multi-stage rotating cascade particle deposition stacking experimental method.

[0068] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A method for a multi-stage rotating cascade particle deposition stepwise superposition experiment, characterized in that, include: S1. The initial three-dimensional profile of the turbine experimental blade was scanned using a white light interferometer to establish a profile registration dataset. The initial mass of the turbine experimental blade and the particle size and density parameters were measured and integrated to form the turbine blade deposition experiment reference data. S2. Based on the turbine blade deposition experiment baseline data, a single-stage rotating blade configuration of the turbine is constructed. The operating speed of the drive motor and the particle injection outlet speed are set. The blades are scanned, registered and differentially calculated to obtain the deposition thickness distribution data of the single-stage turbine blade. S3. Add a stationary blade disk to the single-stage rotating blade cascade configuration of the turbine to form a two-stage rotating blade cascade configuration of the turbine. Scan the registration difference and perform deposition falloff correction to obtain the deposition thickness distribution data of the two-stage turbine blades. S4. Assemble all bladed disks into the turbine two-stage rotating blade cascade configuration to form a turbine full multi-stage rotating blade cascade configuration. Scan the three-dimensional profile of each stage blade and perform deposition shedding correction to obtain the deposition thickness distribution data of the full multi-stage turbine blades. S5. Perform layer-by-layer differential calculations on the deposition thickness distribution data of single-stage, two-stage, and multi-stage turbine blades to separate the independent deposition amount of the second-stage blades and the deposition amount of the third-stage and above blades, and generate a blade differential distribution set. S6. Based on the differential distribution set of the blade, a preliminary region is divided, and the morphology of the preliminary region is observed by scanning electron microscopy and the energy dispersive spectroscopy composition is analyzed to generate blade deposition and erosion zoning distribution data. S7. Based on the blade deposition erosion zoning distribution data, calculate the mass closure error between the three-dimensional scanning mass change and the actual balance mass change. Based on the mass closure error, extract the detachment area data in the blade deposition erosion zoning distribution data. Use the neighborhood equivalence method to correct the detachment area data and verify its accuracy. Combine the independent deposition amount of the second-stage blade and the deposition amount of the third-stage and above blades to generate a quantitative assessment of the service life of the turbine blade. The deposition detachment correction includes: identifying the detachment area based on the joint criteria of thickness variation distribution and scanning electron microscopy morphology, and correcting the deposition thickness of the detachment area using the neighborhood equivalence method; the neighborhood equivalence method includes: using neighboring non-detached grid cells with the same blade surface properties and under the same aerodynamic impact conditions as equivalent references, and estimating the original deposition thickness of the detachment area by weighted averaging.

2. The method of claim 1, wherein, S1 includes: The turbine blade and particles of the aero-engine to be tested were obtained. The reference pit in the non-working area of ​​the turbine blade was used as the registration reference. The initial three-dimensional profile of the turbine blade was scanned by a white light interferometer to establish a profile registration dataset. The initial mass of the turbine blades was determined using a high-precision electronic balance, and the particle size and density parameters were determined using a laser particle size analyzer and a densitometer, respectively. The integrated profile registration dataset, the initial mass of the turbine experimental blade, and the particle size and density parameters are used to form a unified benchmark data for turbine blade deposition experiments.

3. The method of claim 1, wherein, S2 includes: Based on the benchmark data of turbine blade deposition experiments, a first-stage moving blade disk of the corresponding specification was matched, and a single-stage rotating blade cascade configuration of a turbine with only the first-stage moving blade disk installed was built on the turbine blade cascade test bench. Based on the single-stage rotating blade configuration of the turbine, the driving motor speed and particle ejection outlet velocity were set according to the turbine blade deposition experiment benchmark data. The inlet temperature and pressure of the experimental bench were kept at preset constant values, and the single-stage blade particle deposition experiment of preset duration was started. After the experiment, the first-stage moving blade disk was disassembled and loose particles were removed by non-contact purging. The complete three-dimensional profile of each blade was scanned using a white light interferometer. Registration and differential calculations were performed using the profile registration dataset to obtain the deposition thickness distribution data of the single-stage turbine blade.

4. The method of claim 3, wherein, S3 includes: Keeping the test bench foundation and connection structure of the single-stage rotating blade cascade configuration of the turbine unchanged, a stator disk of the corresponding specification is added to form a two-stage rotating blade cascade configuration of the turbine. Based on the two-stage rotating blade configuration of the turbine, the two-stage blade particle deposition experiment was started with the same drive motor operating speed, particle injection outlet velocity, inlet temperature, inlet pressure and experimental duration as the single-stage experiment. After the experiment, the two-stage bladed disks were disassembled in sequence and loose particles were removed by non-contact purging. The complete three-dimensional profile of each stage of the blades was scanned using a white light interferometer. The profile registration dataset was combined with registration difference and deposition shedding correction to obtain the deposition thickness distribution data of the two-stage turbine blades.

5. The method of claim 4, wherein, S4 includes: Keeping the test bench foundation and connecting structure of the two-stage rotating blade cascade configuration of the turbine unchanged, the remaining moving blade disks and stationary blade disks of each stage are assembled in sequence to form a full multi-stage rotating blade cascade configuration of the turbine. Based on the turbine multi-stage rotating blade configuration, the drive motor operating speed, particle injection outlet velocity, inlet temperature, inlet pressure and experimental duration were kept the same as those of the single-stage and two-stage blade particle deposition experiments, and the multi-stage blade particle deposition experiment was started. After the experiment, all bladed disks at each stage were disassembled in sequence and loose particles were removed by non-contact purging. The complete three-dimensional profile of each stage of the blades was scanned using a white light interferometer. The profile registration dataset was combined with registration difference and deposition shedding correction to obtain the deposition thickness distribution data of the entire multi-stage turbine blades.

6. The method of claim 1, wherein, S5 includes: Acquire deposition thickness distribution data for single-stage turbine blades, two-stage turbine blades, and all multi-stage turbine blades, and establish a spatial data matrix of deposition thickness for all multi-stage turbine blades. Based on the spatial data matrix of full multi-stage deposition thickness, the deposition thickness values ​​of the corresponding spatial positions of turbine blades in adjacent configuration levels are calculated by difference to obtain the reverse influence of the newly added turbine blades on the deposition of existing turbine blades. The deposition thickness distribution of the second-stage turbine blade is directly extracted from the deposition thickness distribution data of the two-stage turbine blade to obtain the independent deposition amount of the second-stage blade; the deposition thickness distribution of the third-stage and above blades is directly extracted from the deposition thickness distribution data of the entire multi-stage turbine blade to obtain the deposition amount of the third-stage and above blades. By integrating the reverse influence amount, the independent deposition amount of the second-level blade, the deposition amount of the third-level and above blades, and the corresponding spatial coordinate information, and removing abnormal measurement points and missing data points, a blade differential distribution set containing the spatial coordinates of the blade surface and the corresponding thickness difference value is generated.

7. The method according to claim 1, characterized in that, S6 includes: Based on the differential distribution set of the blades, and combined with the positive and negative values ​​and amplitude range of the thickness difference at each point on the surface of each blade, the surface of each blade is divided into a preliminary region containing the deposition region, the erosion region and the dynamic equilibrium critical region. Scanning electron microscopy morphology observation and energy dispersive spectroscopy analysis were carried out sequentially on the preliminary area to simultaneously acquire surface micromorphology features and elemental composition information under the same field of view; By combining surface micromorphological features and elemental composition information, the depositional regions, erosion regions, and dynamic equilibrium critical regions on the surface of blades at all levels are identified, and the distribution data of blade depositional erosion zones are generated.

8. The method according to claim 1, characterized in that, S7 includes: The measured mass of turbine blades after single-stage, two-stage, and multi-stage blade particle deposition experiments was obtained, and the change in measured mass was calculated in combination with the initial mass of the turbine blades. Based on the distribution data of blade deposition erosion zones and the initial mass of the turbine experimental blade, the three-dimensional scanning mass change is calculated by combining the deposition thickness distribution data of single-stage, two-stage, and multi-stage turbine blades, and the mass closure error between the three-dimensional scanning mass change and the measured mass change by the balance is calculated. Based on the mass closure error, the data of the detachment zone in the blade deposition erosion zoning distribution data is extracted. The detachment zone data is corrected by the neighborhood equivalence method. The accuracy of the corrected data is verified by recalculating the mass closure error, and the corrected blade deposition erosion zoning distribution data is obtained. By combining the independent deposition amount of the second-stage blade, the deposition amount of the third-stage and above blades, and the corrected blade deposition erosion zoning distribution data, blade deposition erosion damage parameters are constructed. Based on the blade deposition erosion damage parameters and the preset failure threshold, a quantitative assessment of the service life of the turbine blade is generated.

9. A multi-stage rotating cascade particle deposition stacking experimental system, used to implement the method as described in any one of claims 1-8, characterized in that, include: The benchmark database module is used to scan the initial three-dimensional profile of the turbine experimental blade using a white light interferometer, establish a profile registration dataset, determine the initial mass of the turbine experimental blade and the particle size and density parameters, and integrate them to form benchmark data for turbine blade deposition experiments. The single-stage deposition module is used to build a single-stage rotating blade configuration of the turbine based on the turbine blade deposition experiment benchmark data, set the operating speed of the drive motor and the particle injection outlet speed, scan the blade and register and perform differential calculations to obtain the single-stage turbine blade deposition thickness distribution data. The two-stage deposition module is used to add a stator disk to the single-stage rotating blade cascade configuration of the turbine to form a two-stage rotating blade cascade configuration of the turbine. It scans and registers the difference and performs deposition loss correction to obtain the deposition thickness distribution data of the two-stage turbine blades. The full-stage deposition module is used to assemble all bladed disks into the turbine two-stage rotating blade cascade configuration to form a turbine full multi-stage rotating blade cascade configuration, scan the three-dimensional profile of each stage blade and perform deposition shedding correction, and obtain the deposition thickness distribution data of the full multi-stage turbine blades. The deposition separation module is used to perform layer-by-layer differential calculations on the deposition thickness distribution data of single-stage, two-stage, and multi-stage turbine blades, separating the independent deposition amount of the second-stage blades and the deposition amount of the third-stage and above blades, and generating a blade differential distribution set. The partitioning analysis module is used to divide the preliminary regions based on the differential distribution set of the blades, perform scanning electron microscopy morphology observation and energy dispersive spectroscopy analysis on the preliminary regions, and generate blade deposition and erosion partitioning distribution data. The service life assessment module is used to calculate the closure error between the three-dimensional scanning quality change and the measured quality of the balance based on the blade deposition erosion zoning distribution data, extract the detachment zone data in the blade deposition erosion zoning distribution data based on the closure error, correct the detachment zone deposition data using the neighborhood equivalence method and verify the accuracy, and generate a quantitative assessment of the turbine blade service life by combining the independent deposition amount of the second-stage blade and the deposition amount of the third-stage and above blades.

10. A computer-readable storage medium storing instructions thereon, characterized in that, When the instruction is executed by the processor, it implements a multi-stage rotating cascade particle deposition stacking experimental method as described in any one of claims 1-8.

Citation Information

Patent Citations

  • Thermal barrier coating deposition characteristic numerical simulation prediction method considering surface roughness

    CN120473052A

  • Sensing systems and methods for building an intelligent model of particulate ingress detection in turbine systems

    US20230029650A1