Blade Crack Research Articles

Damage evaluation and life assessment of steam turbine blades

Steam turbines in their last stages, L-0 stage, generate approximately 12% of the total power produced by thermoelectric plants. These steam turbines blades present centrifugal loads and unstable flows by rotating. Both there are the main causes for fatigue crack initiation [1]. As a consequence of this crack initiation, dynamical behavior in blades is altered. These dynamical variations in damping, stiffness, and natural frequencies in presence of cracks are investigated and evaluated to prevent damage and life assessment (useful life) until to failure. Experimental fatigue tests and modal analysis in samples of 410SS, stainless steel were performance to obtain the S-N curve and boundary conditions for the numerical analysis performance in Ansys workbench®. These experimental tests provide some magnitudes to numerical simulation. Numerical modal analysis, random vibrations with power spectral density are performed to detect these changes in damping, stiffness, and natural frequencies in the group of ten blades of the last stage of a 110 MW steam turbine. The Steinberg 3-band technique and Palmgren-Miner cumulative damage rule were used to calculated fatigue damage and useful life. Results show that different crack sizes change the performance of steam turbine blades. If changes in damping, stiffness, and natural frequencies with fractured blades are detected, it is possible to determine these damages and predict useful life. The cumulative damage is obtained with RD = 1.232 and useful life is estimated in the blades group with changes in damping ratio due to cracks in blades with a result of 6.14 min.

Influences of blade crack on the coupling characteristics in a bladed disk with elastic support

In aero-engines, the blades are exposed to severe environments and subjected to complex conditions, which can cause damage and cracks. Existing analytical models focus primarily on the blade crack in a single blade. Even in the cracked-bladed disk system, most theoretical models are based on the rigid disk assumption, and few studies pay attention to the boundary conditions of the disk. A new model is proposed in this paper to compensate for deficiencies in existing theoretical models. The flexible disk and blade are deduced based on Kirchhoff plate and Timoshenko beam theories. The crack-caused stiffness loss is calculated through the released strain energy. In addition, the elastic support on the disk's inner edge is simulated by the linear spring stiffness and torsional stiffness. The accuracy of the developed model is verified by comparing it with the finite element method and experimental test. After that, the influences of the crack depth and crack location on the coupling characteristics of the system are systematically investigated. The results indicate that blade crack mainly affects the blade-predominated modes, leading to the mistuned disk mode shapes, which are symmetrically distributed along the line between the cracked blade and the disk center. Moreover, as the blade crack becomes severe or approaches the blade root, the first natural frequency of the paired orthogonal modes decreases; the first mode shape of the paired tuned high nodal diameter degenerates into mistuned low nodal diameter; the new coupling characteristics between the mistuned disk and elastic support become more stronger. The proposed methodology can serve as a theoretical basis for diagnosing faults and monitoring safety in the bladed disk coupling system.

Blade Crack Diagnosis Based on Blade Tip Timing and Convolution Neural Networks

The diagnosis of blade crack faults is critical to ensuring the safety of turbomachinery. Blade tip timing (BTT) is a non-contact vibration displacement measurement technique, which has been extensively studied for blade vibration condition monitoring recently. The fault diagnosis methods based on deep learning can be summarized as studying the internal logical relationship of data, automatically mining features, and intelligently identifying faults. This research proposes a crack fault diagnostic method based on BTT measurement data and convolutional neural networks (CNNs) for the crack fault detection of blades. There are two main aspects: the numerical analysis of the rotating blade crack fault diagnosis and the experimental research in rotating blade crack fault diagnosis. The results show that the method outperforms many other traditional machine learning models in both numerical models and tests for diagnosing the depth and location of blade cracks. The findings of this study contribute to the real-time online crack fault diagnosis of blades.

Numerical investigation of crack initiation in high-pressure gas turbine blade subjected to thermal-fluid-mechanical low-cycle fatigue

A thermo-fluid-structural analysis procedure was proposed in this study to investigate the lifespan of a realistic cooling gas turbine blade. The temperature and pressure data extracted from computational fluid dynamics simulations were applied as the loading force to calculate the strain and stress using the finite element (FE) method. Subsequently, a fatigue analysis was conducted to estimate the blade life and predict the position of cracks. Here, the fatigue model used to calculate the blade life and damage was based on the strain-life model. The results showed that although the cooling film covered the entire blade surface, inefficient cooling was still observed in some regions, such as the tip and platform, which lead to blade overheating. A high pressure distribution of approximately 1.374 MPa was formed primarily on the leading tip (LT) of the pressure side (PS) blade; this pressure decreased toward the suction side. The FE simulations revealed that temperature caused a high stress at the tip and platform regions, whereas pressure caused stress at the leading edge fillet and trailing edge (TE) root slot. Thermal stress accounted for a greater amount of stress on the blade in comparison to the pressure force. Moreover, the combination of thermal and pressure stresses increased the stress at the TE root slot. Although the maximum stress in almost all the regions was lower than the ultimate strength, cracks still appeared under low-cycle start–stop loading. According to the fatigue life predictions, cracks would appear on the TE root slot, middle-tip, trailing tip, LT, and on the PS edge platform after approximately 150, 484, 701, 673, and 374 start–stop cycles, respectively. These fatigue life estimations showed good agreement with the blade crack patterns.

Dynamic modeling of rotary blade crack with regard to three-dimensional tip clearance

To investigate the dynamic characteristics of the three-dimensional blade tip clearance (3D-BTC) of rotary blade cracks, a novel dynamic model of the rotary blade crack with regard to the 3D-BTC is derived based on the Timoshenko beam theory. In this model, the modal shape functions of the blade are derived based on the transfer matrix method, and the axial and circumferential aerodynamic loads are introduced to better obtain the dynamic responses of the blade vibration. Then, a breathing crack model of the blade trailing edge crack is established by considering the coupling effects of the centrifugal stress, axial bending stress, and circumferential bending stress. The crack opening width is introduced into the breathing crack model to comprehensively depict the crack together with the crack depth and crack location. Moreover, a numerical calculation method for the 3D-BTC is proposed for the rotary blade with cracks. The dynamic model is validated by comparing it with the finite element model and the experiment. Finally, the effects of the blade trailing edge crack on the 3D-BTC are analyzed. The results show that the cracked blade's first-order natural frequency decreases with the crack depth increase, and it increases first and then decreases slightly when the crack location moves away from the blade root. The effects of crack opening width on the first-order natural frequency are relatively small due to its small size. In the meantime, the nonlinear damage indicators (NDIs) of the 3D-BTC are calculated. The results show that the NDIs can reflect the severity of the blade crack, and the proposed model can comprehensively analyze the effects of rotary blade crack on the 3D-BTC, which is beneficial for the diagnosis method for blade cracks based on the 3D-BTC.

Infrared Thermal Imaging-Based Turbine Blade Crack Classification Using Deep Learning

Non-destructive testing is widely applied for the detection and identification of defects in turbine blades of modern aircraft engines. Cracks in turbine blades can affect the turbine performance and pose a risk to safety and service life. For Original Equipment Manufacturers it is, therefore, essential to be able to identify all defects. Heat flow thermography offers, compared to the often used penetrant testing, the potential to improve the detection of defects in turbine blades and is contact-free, reproducible, quick to apply, and can be automated. With induction (heat flow) thermography, it is even possible to detect cracks that lie below the surface and therefore are not externally visible. However, manual inspection of thermography images is very time-consuming. By automating the image classification procedure with a deep learning technique, the speed and accuracy of the classification can be improved over a manually performed classification. The development objective of this AI application is expected to support and assist the highly skilled and experienced inspection specialists in the medium term. Our solution is based on convolutional neural networks. Several challenges of the AI training process, including data imbalance, a small dataset, and extremely small cracks in large images are addressed.

Dynamic Characteristics Analysis of 3D Blade Tip Clearance for Turbine Blades with Typical Cracks

As well known, the crack is a typical fault of turbine blades, which is difficult to detect due to the weak fault features contained in the rotor vibration. Although some fault diagnosis methods based on blade tip timing (BTT) are proposed to detect blade cracks, the limited fault features can be extracted because the BTT signals contain only one-dimensional fault information. Based on the dynamic analysis, the cracked blade undergoes three-dimensional (3D) deformation so that it would be reflected in the changes of the three-dimensional blade tip clearance (3D-BTC). Therefore, the dynamic characteristics of the 3D-BTC of turbine blades with typical cracks should be investigated to obtain its change rule and achieve a more comprehensive diagnosis of blade cracks. In this paper, the 3D finite element model of the turbine blisk is established, and the fluid-thermal-solid coupling analysis is implemented to calculate the blade deformation. Then, a novel dynamic computation method is proposed to calculate the 3D-BTC based on the blade deformation. The typical blade cracks are taken into consideration, and the dynamic characteristics of the 3D-BTC regarding the typical cracks are analyzed to obtain its change rule. The results show that abundant crack fault information can be obtained based on the 3D-BTC, which can facilitate the comprehensive diagnosis of turbine blade cracks.

Analysis of Failure Causes of 0Cr19Ni9 Blade Crack

In this paper, fracture morphology observation, microstructure analysis, standard corrosion test, and other experimental methods are used to analyze the causes of corrosion and cracking of steam turbine blades (0Cr19Ni9) produced by a certain company. The results showed that cracks appeared during the operation of the blade. A large number of dimples were observed under the scanning electron microscope, almost all of which were shear dimples; at the same time, fatigue bands were also distributed in them. The internal cause of the cracks is the long-term service, which causes the carbon and the alloy to undergo a diffusion reaction to form carbon-rich areas and carbides. They are thermally corroded in the service environment, and the carbon-rich zone is reduced, which reduces the strength of the zone; carbides accumulate at the grain boundary, causing groove-shaped defects at the grain boundary to form a cutting effect. It is easy to form a source of fatigue cracks. The external cause is that the material is subjected to transverse stress for a long time during operation, which leads to the generation of cracks. This article analyzes the failure causes of the blade and proposes related solutions, which have important reference value for production practice.

Two-level fusion of multi-sensor information for compressor blade crack detection based on self-attention mechanism

Nowadays, acoustic and vibration signals have been widely used to detect crack and its degree on compressor blades for non-destructive evaluation and structural health monitoring. Due to complex working conditions and interference of strong noise, single signal-based methods reach unsatisfactory accuracy. To improve the reliability and accuracy of crack detection, the two-level fusion method based on Mahalanobis distance (MD) and self-attention mechanism is proposed based on multi-acoustic, vibration, and acoustic emission information. First, the MD is calculated to measure the similarity between raw samples and group, which can fuse the same type of samples in data level. Then, the processed raw and data-level fusion samples are inputted to one-dimensional convolutional neural network, and the preliminary decisions are obtained. Finally, the decision-level fusion based on self-attention mechanism and credit assignment is proposed to modify and correct the preliminary decisions. The compressor experiments are implemented to test the proposed method under single and mixed working conditions with strong noise intensity. Further, the results illustrate that the proposed method can accurately detect crack for compressor blades. By comparing with other data fusion approaches, the advantage of the proposed method is validated under complex conditions.

RESEARCH AND OPTIMIZATION OF REFURBISHMENT OF HPT BLADES OF THE D-18T AIRCRAFT GAS TURBINE ENGINE BY MICRO-PLASMA POWDER WELDING

In current work, peculiarities of the micro-plasma powder welding deposition process applied to the batch refurbishment of D-18T aircraft engine HPT blades made of ZhS32-VI nickel superalloy with limited weldability have been studied. It has been demonstrated that extending operating resource over 6-8 thousand hours leads to an increase in ZhS32-VI “base-deposited metal” weld’s cracking susceptibility during welding and subsequent thermal processing operations. It has been shown that providing stable forming of deposited bead on the shroud edge’s surface requires applying nonstationary impulse modes of straight polarity welding current. Considering the significant amount of technological parameters of the process studied, subjective analysis of such welding modes is extremely complicated. We have introduced the method of specifying requirements for such single-layer micro-plasma powder welding deposition modes with welding current in a range of 7-20 A according to the criteria of effective arc heating power and heat input, which involves using the system for registration and digital processing of welding current-welding bead deposition time dependency. Based on the analysis of statistical data on the quantitative evaluation of cracking susceptibility of the investigated weld, the optimal range of average values has been discovered for these generalized welding deposition mode parameters to be applied in a manual or automated process, which provides no more than several percent of cracked blades detected at the end of refurbishment technological cycle. We have shown that significant technological parameters, which affect the amount of technological defects during mentioned blades’ refurbishment, are the average value of effective welding current and ZhS32 superalloy powder quality, primarily determined by oxygen and nitrogen average weight content in its dispersed particles. It was also shown that the increase in average weight content of gas impurities in the filler powder, primarily oxygen, causes a significant increase in energy consumed on deposited bead formation during the micro-plasma powder deposition process. An initial quality control method for the ZhS32-VI filler powder range has been proposed. The method is based on the average weight content of oxygen and nitrogen, evaluated by reducing fusion in transporting gas flow and on evaluation criteria of effective arc heating power and heat input average values of witness sample blades’ refurbishment process and their conformity to previously defined optimal. An evaluation of future application possibilities for registration systems and deposition mode analysis in batch repair conditions of nickel superalloy aircraft parts has been introduced.

Axial-bending coupling vibration characteristics of a rotating blade with breathing crack

Blade crack is one of the most common blade failures of rotating machines which may lead to catastrophic accidents. Although many nonlinear vibrations of breathing crack have been investigated, there are few reports on the axial-bending coupling vibration caused by crack. With the aim of providing physical insight into the mechanisms of axis-bending coupling due to crack, a novel axial-bending coupled breathing crack model (ABCBCM) for the rotating blade is proposed in this study. The proposed ABCBCM is analytically formulated based on the Timoshenko beam theory and Castigliano’s principle, and validated by comparing the natural and dynamic characteristics with the finite element model (FEM) and experimental tests. In addition, the effects of boundary conditions (rotational speed), crack parameters (crack depth and crack location) and loading conditions (excitation load) on the vibration characteristics of the cracked blade are investigated. The results indicate that the blade axial response is more sensitive to the nonlinearity caused by the breathing crack than the bending response; the axial-bending coupling caused by the crack changes the equilibrium position of the axial and bending displacement; the phase portrait of axial response and the orbit of axial-bending displacement are effective methods to detect breathing crack; and the axial displacement amplitude ratio and the axial acceleration amplitude ratio also are valuable indicators to estimate the severity of the blade crack. The proposed model can provide valuable insights for fault diagnosis and safety monitoring of cracked blade.

Blade tip timing for monitoring crack propagation of rotor blades using Block-AOLS

Blade tip timing (BTT) as a non-contact measurement technique can identify the blade vibration parameters. However, the signal recorded by BTT usually does not satisfy the Nyquist sampling theorem. To monitor blade crack propagation, it is important to accurately reconstruct the BTT signal to track the natural frequency shift of the rotor blade. In this paper, a sparse reconstruction method named block-accelerated orthogonal least-squares (Block-AOLS) is proposed to reconstruct the undersampled BTT signal in the time-frequency domain. Block-AOLS not only inherits the advantage of the high accuracy of AOLS but also makes use of the characteristic of block sparsity to reduce the interference components in the BTT signal. Furthermore, Block-AOLS is used to reconstruct the undersampled BTT signal to track the natural frequencies of cracked blades. The advantage of Block-AOLS is illustrated through simulation by comparing with the classical orthogonal matching pursuit. In the spin testing, the vibration of a blisk with eight blades is measured by both BTT and strain gauges, where two symmetrical blades without artificial cracks suddenly broke. The process of crack propagation can be inferred according to the shift of natural frequencies between the perfect and cracked blades. Moreover, a vibration frequency prediction method is provided to avoid comparing the natural frequencies among different blades for crack propagation monitoring, which is suitable to monitor the mistuning blades. To verify the reliability of the crack propagation monitoring using BTT, the strain signal with interval sample is analyzed by synchroextracting transform. Compared the natural frequency of the perfect blade with that of the cracked blade, the time when the natural frequency of the cracked blade shift is the same as that of the BTT result.

Blade crack detection using variational model decomposition and time-delayed feedback nonlinear tri-stable stochastic resonance

For centrifugal fans, the emergence of amplitude modulation (AM) components is a pivotal feature for blade crack fault. However, AM features extraction are challenging under strong interferences. A novel method is proposed to solve this problem using variational mode decomposition (VMD) and time-delayed feedback nonlinear tri-stable stochastic resonance (TFNTSR). Firstly, an improved degree of cyclostationarity (IDCS) index is constructed for parameters optimization. Then, intrinsic mode functions (IMFs) are obtained using the VMD method, and one of them which has the maximum IDCS value is selected. Finally, the selected IMF is further processed by the TFNTSR method, thereby completing the blade crack detection. Verified by simulations and experiments, VMD-TFNTSR method has a better performance on blade crack detection.

Data and Decision Level Fusion-Based Crack Detection for Compressor Blade Using Acoustic and Vibration Signal

Crack detection is important for compressor blades. Due to the interference of strong noise, the existing methods based on single signal reach unsatisfactory performance. To improve the reliability and accuracy, a novel crack detection method is proposed by fusing the information from acoustic and vibration signals (AVS) in data and decision level. Firstly, the data-level fusion method is proposed by calculating Hoyer and improved cosine similarity, which is applied for vibration signals fusion based on its sparsity and similarity. Then, the raw AVS and data-level fusion samples are trained in the one-dimensional convolutional neural network (1D CNN), and the preliminary classification is obtained based on the probabilities of different cracks. Finally, the decision-level fusion method with changed reliability assignment is proposed through modifying and correcting the preliminary results of 1D CNN, which can reach reliable decisions and realize crack detection for compressor blade. The proposed method is tested by compressor experiments with three cracked blades under four single and one mixed working conditions. The results illustrate that the proposed method can make full use of AVS and detect cracks reliably under five working conditions. Furtherly, the superiority of the proposed method is validated by comparing with other crack detection approaches.

Investigating Rotor Conditions on Wind Turbines Using Integrating Tree Classifiers

Renewable wind power is productive and feasible to manage the energy crisis and global warming. The wind turbine’s blades are the essential components. The dimension of wind turbine blades has been increased with blade sizes varying from approx. 25 m up to approx. 100 m or even greater with a specific purpose to increase energy efficiency. While wind turbine blades tend to be highly stressed by environmental conditions, the wind turbine blade must be constantly tested, inspected, and monitored for wind turbine blades safety monitoring. This research presents a methodology adaptation on machine learning technique for appropriate classification of different failure conditions on blade during turbine operation. Five defects were reported for the diagnosis study of defective wind turbine rotor blades, and the considered defects are blade crack, erosion, loose hub blade contact, angle twist, and blade bend. The statistical features have been drawn from the recorded vibration signals, and the important features was selected through J48 classifier. Eight tree-dependent classifiers were used to categorize the state of the rotor blades. Among the classifiers, the least absolute deviation tree performed better with the classification percentage of 90% ( Kappa statistics = 0.88 , MAE = 0.0362 , and RMSE = 0.1704 ) with a computational time of 0.06 s.

Reconstruction of fan blade dynamic loading prior to its fracture

To maintain the functionality of fan blades, it is important to know the distribution of dynamic stresses in the blade, their amplitude and vibration frequency. Understanding of the dynamic loading pattern will allow us to determine under what conditions the engine was operated, to identify and prevent emergencies that could lead to the blade fracture. The purpose of this work is to understand the cause of the fan blade fracture that occurred during the engine ground start. Due to fractographic analysis of blade fragments it was revealed that the fracture occurred due to the initiation of fatigue cracks in blades. The place of crack initiation and parameters of crack growth were established, spectral analysis of the fracture was carried out. To establish the reason of fatigue crack initiation it was necessary to determine the dynamic state of the blades during their destruction. The sections of the second stage of stable crack growth, during which fatigue striations are formed, were determined using the fractographic method. Using the spacing of the fatigue striations and Paris's law, the crack stress intensity range was determined. Modeling of crack propagation in the blade was carried out to define the stress state. The ability to determine the stress intensity factor at each step of crack growth and its comparison with research data made this work possible. The simulation showed under what conditions manifold increase in stresses occurs and made it possible to obtain the expected value of vibration amplitude. Additional modal analysis showed a resonant form that caused the fatigue nature of crack propagation. The demonstrated approach established crack growth conditions and revealed the cause of blade fracture.

Fault feature recognition of centrifugal compressor with cracked blade based on SNR estimation and adaptive stochastic resonance

The centrifugal compressor is widely used in modern industry, and the blades are prone to crack due to complex loads and working conditions. Owing to the fault features caused by initial blade cracks are week and easily interfered by strong noise, it is difficult to recognize accurately by traditional methods. In this study, the SNR estimation and adaptive stochastic resonance (SEASR) method is proposed for fault feature recognition of centrifugal compressor with cracked blade. Based on the relationship between SNR and stochastic resonance (SR) parameters, the subtract noise by empirical mode decomposition (SNEMD) method is established with second-order and fourth-order moment (M2M4), hence solving the problem of adaptive SR. The effectiveness of SEASR is tested and analyzed by simulation signals and experimental data. The results demonstrate that the proposed method can accurately recognize fault features of cracked blades and compound faults, where the output SNR is superior to that of the previous methods. It is a new method to realize fault diagnosis for centrifugal compressor with cracked blade and other rotating machinery.

Mechanical property analysis of shaft tubular pump rotor with different blade angle based on fluid-structure interaction

Abstract Shaft tubular pump is widely used in large and medium-sized pumping stations with ultra-low head. But occasionally the blade cracks or even breaks, which seriously restricts the safe and stable operation of the pump station. In order to obtain the mechanical properties of the pump rotor with different blade installation angles, the fluid-structure interaction numerical simulation method is applied to analyze the structure of the pump rotor. Based on the SST k-ω model and the structured grid of the whole flow field, the three-dimensional numerical simulation of the internal flow field of the pump device is carried out by using CFX, and the reliability of the simulation is verified by model test. Then, the static analysis and modal analysis of the pump rotor structure under different working conditions are carried out by using Workbench. From the results, it can be found that the axial force increases with the increase of blade installation angle at a same flow rate. And at a same flow rate, the maximum deformation and equivalent stress increases with the increase of the blade installation angle. Finally, the change of the blade installation angle mainly affects the third and sixth order frequency which are only showed as blade vibration. Therefore, all of these results indicate that the blade installation angle has obvious influence on the mechanical property of the rotor of shaft tubular pump.

Digital Twin-Driven Crack Monitoring for Rotating Blade: An L1 regularization Method

Abstract Crack-related damage of rotating blade is one of the most common failures of aeroengines. However, conventional blade crack monitoring methods are challenging to obtain the exact location and severity of the crack. In this study, a digital twin-driven blade crack monitoring method using the L1 regularization and sensitivity-based model updating is proposed to quantify the location and severity of crack damage for rotating blade. First and foremost, a digital twin model framework for rotating blade is established, including physical entity, digital entity, and the connection between them. Then a digital entity based on finite element model and an L1 regularized objective function based on the residual of natural frequencies are constructed, where the damage parameter of crack could be obtained by using the sensitivity-based model updating technology, hereby the location and severity of crack could be determined. Finally, numerical simulations are carried out, which verify that the proposed method is capable of accurately identifying the location and severity of crack, and could provide a reference for further research in this field.

Perancangan Aktivitas Pemeliharaan dengan Metode Reliability pada Sistem Main Rotor Blade Helikopter BELL 412EP Studi Kasus Penerbad Semarang

BENGPUSPENERBAD is a central workshop specifically owned by the Indonesian Army in Semarang, where aircraft and helicopters are subject to preventive and corrective maintenance. During the period 2018-2020 the BELL 412 EP helicopters often experienced unscheduled maintenance which resulted in losses resulting in flight failure, this was due to a malfunction in the Main Rotor Blade system. The malfunction of the Main Rotor Blade system is dominated by the hub component, namely on the shaft associated with Crack, Noise and Burn damage, and blade components, namely Corrosions, Nicks, Scratches, Dents, Deformation Holes and Crack. Therefore, the aim of this study is to eliminate failure system and unscheduled maintenance.
 To eliminate failure and unscheduled maintenance, the authors perform an analysis with the FMEA method and calculate using the distribution method, then design the appropriate activities on the system. The data used for analysis are qualitative and quantitative data, where the authors use data from Time To Failure (TTF) and Time To Repair (TTR).
 According to calculations using the current distribution we found that the Main Rotor Blade system was on time during the 200 operational flight hours, for which the reliability figure was 0.70. Maintenance activities in accordance with the operational conditions of the main rotor blade operating system are preventive maintenance with activities for tool preparation, inspection or chek, dismantling and servicing, and finally installation and finishing activities.