A system and method for detecting thermal stress in aero-engine blades
By using a collaborative detection system of green laser and near-infrared laser, multi-modal detection of thermal stress in aero-engine blades was achieved, solving the problem of limited detection depth and efficiency in existing technologies and improving detection capabilities under complex operating conditions.
Patent Information
- Application Number
- CN202510827832.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2025-06-19
- Publication Date
- 2026-11-10
- Estimated Expiration
- 2045-06-19
AI Technical Summary
Existing technologies for detecting thermal stress in aero-engine blades can only detect a single mode, limiting the depth and efficiency of detection and failing to meet the testing requirements under complex operating conditions.
By employing a green laser and a near-infrared laser working in tandem and combining the beams through a beam splitter, along with a scanning control unit and a signal receiver, synchronous acquisition and processing of ultrasonic and photoacoustic signals are achieved, generating detection results including maximum stress value, crack coordinates, and safety factor.
It enables thermal stress detection of the entire thickness of aero-engine blades, adapts to real-time monitoring under complex operating conditions, and improves blade life and flight safety assurance capabilities.
Smart Images

Figure CN120760903B_ABST
Abstract
Description
Technical Field
[0001] This disclosure relates to the field of aero-engine blade manufacturing and testing technology, and in particular to a system and method for detecting thermal stress in aero-engine blades. Background Technology
[0002] Aero-engine blades generate thermal cycling fatigue internal stress under long-term high temperature, high pressure, and high speed operating conditions. This internal stress is mainly manifested as the superposition effect of thermal stress, residual stress, and phase transformation stress. The release of internal stress can lead to many problems. For example, the release of internal stress during electroplating can cause uneven deposition of metal ions, resulting in large-area peeling of the coating. Therefore, it is necessary to detect the thermal stress of engine blades.
[0003] In related technologies, laser detection technology is usually used to detect the thermal stress of engine blades. However, the detection technology in these technologies can only detect thermal stress information of a single mode, and the detection depth and efficiency are limited, which cannot meet the detection requirements of thermal stress of aero-engine blades under complex working conditions. Summary of the Invention
[0004] This disclosure provides a system and method for detecting thermal stress in aero-engine blades.
[0005] According to a first aspect of this disclosure, a system for detecting thermal stress in aero-engine blades is provided. The system includes: a green laser, a near-infrared laser, and a beam splitter, a scanning control unit, a signal receiver, and a processing module connected in sequence; the green laser and the near-infrared laser are respectively connected to the beam splitter.
[0006] The green laser is used to output a first laser beam to the engine blades;
[0007] The near-infrared laser is used to output a second laser beam to the engine blades;
[0008] The beam splitter is used to combine the first laser beam and the second laser beam to obtain a composite beam;
[0009] The scanning control unit is used to control the synthetic beam to scan the engine blades;
[0010] The signal receiver is used to receive ultrasonic signals and photoacoustic signals generated by the engine blades based on the synthetic beam;
[0011] The processing module is used to process the ultrasonic signal and the photoacoustic signal to generate a detection result; wherein the detection result includes at least: the maximum stress value of the engine blade, the crack coordinates, and the safety factor.
[0012] Furthermore, the scanning control unit includes: a two-dimensional galvanometer and a six-axis robotic arm;
[0013] The two-dimensional galvanometer is used to reflect the synthesized beam to generate the reflected beam;
[0014] The six-axis robotic arm is used to drive the two-dimensional galvanometer to move, so that the reflected light beam scans the engine blades according to a preset scanning trajectory.
[0015] According to a second aspect of this disclosure, a method for detecting thermal stress in aero-engine blades is provided, applied to the aforementioned aero-engine blade thermal stress detection system, the method comprising:
[0016] Control the green laser to output a first laser beam to the engine blades, and control the near-infrared laser to output a second laser beam to the engine blades;
[0017] The first laser beam and the second laser beam are combined to obtain a composite beam;
[0018] Control the composite beam to scan the engine blades;
[0019] Receive ultrasonic and photoacoustic signals generated by the engine blades based on the synthetic beam;
[0020] The ultrasonic signal and the photoacoustic signal are processed to generate a detection result; wherein the detection result includes at least: the maximum stress value of the engine blade, the crack coordinates, and the safety factor.
[0021] Further, the processing of the ultrasonic signal and the photoacoustic signal to generate a detection result includes:
[0022] The implicit field characteristics of the engine blade are obtained based on the ultrasonic signal and the photoacoustic signal; wherein, the implicit field characteristics are used to characterize the physical state of stress, temperature and material response at each point inside the engine blade;
[0023] The stress tensor of the engine blade is determined based on the implicit field characteristics; wherein, the stress tensor is used to characterize the magnitude, direction and thermo-mechanical coupling state of the stress borne by any point inside the engine blade in three-dimensional space;
[0024] A three-dimensional stress field distribution cloud map of the engine blade is established based on the stress tensor, and the detection result is determined based on the three-dimensional stress field distribution cloud map.
[0025] Furthermore, the step of obtaining the implicit field characteristics of the engine blade based on the ultrasonic signal and the photoacoustic signal includes:
[0026] The ultrasonic propagation delay is determined based on the ultrasonic signal;
[0027] Calculate the elastic modulus and strain rate sensitivity coefficient of the engine blade;
[0028] The temperature gradient of the engine blades is obtained based on the photoacoustic signal, and the coefficient of thermal expansion is calculated based on the temperature gradient.
[0029] The ultrasonic wave propagation delay, temperature gradient, elastic modulus, thermal expansion coefficient, and strain rate sensitivity coefficient are input into a dynamic hash coding network model to obtain an implicit neural radiation field; wherein, the implicit neural radiation field is used to characterize the spatiotemporal evolution of the stress, temperature, and strain multiphysics coupling state of the engine blade in three-dimensional space.
[0030] The implicit field characteristics of the engine blades are determined based on the implicit neural radiation field.
[0031] Further, determining the stress tensor of the engine blade based on the implicit field characteristics includes:
[0032] The implicit field features are input into a multilayer perceptron network, and physical constraint equations are embedded in the multilayer perceptron network to obtain a modified multilayer perceptron network; wherein, the physical constraint equations include at least: momentum conservation equation and energy conservation equation;
[0033] The stress tensor of the engine blade is determined based on the modified multilayer perceptron network.
[0034] Furthermore, the step of establishing a three-dimensional stress field distribution cloud map of the engine blade based on the stress tensor includes:
[0035] The stress tensor is mapped to a continuous three-dimensional stress field function by differentiable interpolation;
[0036] The continuous three-dimensional stress field function is optimized by using a loss function to obtain the optimized three-dimensional stress field function.
[0037] A three-dimensional stress field distribution cloud map is established based on the optimized three-dimensional stress field function.
[0038] Furthermore, determining the detection result based on the three-dimensional stress field distribution cloud map includes:
[0039] The three-dimensional stress field distribution cloud map is dynamically rendered through an interactive interface to obtain a visual rendering result; wherein, the visual rendering result is a color gradient display of the stress gradient of the engine blade;
[0040] The detection results are determined based on the visualization rendering results; wherein, the detection results include at least: the maximum stress value of the engine blade, the crack coordinates, and the safety factor.
[0041] Furthermore, after establishing a three-dimensional stress field distribution cloud map of the engine blade based on the stress tensor, the method further includes:
[0042] The local stress intensity of the engine blades is determined based on the three-dimensional stress field distribution cloud map.
[0043] The dynamic hash resolution of the dynamic hash coding network model is updated based on the local stress intensity to obtain the updated hash coding network model.
[0044] The updated three-dimensional stress field distribution cloud map is determined based on the updated hash-encoded network model.
[0045] Furthermore, the method also includes:
[0046] The reference stress, current stress field, strain rate, and process random noise of the engine blade are processed by a long short-term memory network to obtain the stress field of the engine blade at the next moment.
[0047] The gradient distribution of the engine blades is determined based on the stress field at the next moment;
[0048] The scanning command for the six-axis robotic arm is determined based on the gradient distribution; wherein, the scanning command is used to drive the six-axis robotic arm to move the two-dimensional galvanometer, so that the reflected light beam performs a high-density scan on the high-stress gradient region of the engine blade.
[0049] According to a third aspect of this disclosure, an electronic device is provided. The electronic device includes a memory and a processor, wherein the memory stores a computer program, and the processor executes the program to implement the method described above.
[0050] According to a fourth aspect of this disclosure, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the methods described above.
[0051] According to a fifth aspect of this disclosure, a computer program product is provided. The computer program product includes a computer program that, when executed by a processor, implements the methods described above in this disclosure.
[0052] This disclosure provides a system and method for detecting thermal stress in an aero-engine blade. The system includes: a green laser, a near-infrared laser, and a beam splitter, a scanning control unit, a signal receiver, and a processing module connected in sequence. The green laser and the near-infrared laser are respectively connected to the beam splitter. The green laser outputs a first laser beam to the engine blade. The near-infrared laser outputs a second laser beam to the engine blade. The beam splitter merges the first and second laser beams to obtain a composite beam. The scanning control unit controls the composite beam to scan the engine blade. The signal receiver receives ultrasonic and photoacoustic signals generated by the engine blade based on the composite beam. The processing module processes the ultrasonic and photoacoustic signals to generate detection results. The detection results include at least: the maximum stress value of the engine blade, crack coordinates, and a safety factor.
[0053] As described above, the aero-engine blade thermal stress detection system provided by this disclosure uses a green laser and a near-infrared laser to generate a composite beam through a beam splitter. This composite beam, controlled by a scanning control unit, can achieve precise scanning of complex curved surfaces. Simultaneously, a signal receiver captures ultrasonic and photoacoustic signals, and a processing module further analyzes and processes these signals to generate detection results including the maximum stress value, crack coordinates, and safety factor of the engine blade. This aero-engine blade thermal stress detection system overcomes the limitations of single-mode detection in traditional technologies, achieving simultaneous acquisition of ultrasonic and photoacoustic signals. This allows for detection depth coverage of the entire blade thickness, avoiding the shortcomings of traditional technologies that can only detect the surface layer. Furthermore, by controlling the composite beam to scan the engine blade through the scanning control unit, this system can adapt to the real-time monitoring needs of complex operating conditions such as aero-engine start-up, shutdown, and variable loads, thereby improving the lifespan of aero-engine blades and flight safety assurance capabilities. Attached Figure Description
[0054] The above and other objects, features, and advantages of this disclosure will become more apparent from the more detailed description of the embodiments thereof in conjunction with the accompanying drawings. The drawings are provided to further illustrate the embodiments of this disclosure and form part of the specification. They are used together with the embodiments of this disclosure to explain the disclosure and do not constitute a limitation thereof. In the drawings, the same reference numerals generally represent the same components or steps.
[0055] Figure 1 A schematic diagram of the structure of an aero-engine blade thermal stress detection system provided as an exemplary embodiment of this disclosure;
[0056] Figure 2 A schematic diagram of the structure of an aero-engine blade thermal stress detection system provided as another exemplary embodiment of this disclosure;
[0057] Figure 3 One of the flowcharts for a method of detecting thermal stress on an aero-engine blade provided as an exemplary embodiment of this disclosure;
[0058] Figure 4 A second flowchart of a method for detecting thermal stress in aero-engine blades provided as another exemplary embodiment of this disclosure;
[0059] Figure 5 A flowchart of a method for detecting thermal stress on aero-engine blades provided as another exemplary embodiment of this disclosure;
[0060] Figure 6 A flowchart illustrating an exemplary embodiment of the present disclosure of a method for detecting thermal stress in aero-engine blades.
[0061] Figure 7 A structural block diagram of an electronic device provided as an exemplary embodiment of this disclosure;
[0062] Figure 8 A structural block diagram of a computer system provided as an exemplary embodiment of this disclosure;
[0063] Figure 9 A structural block diagram of a computer program product provided for an exemplary embodiment of this disclosure. Detailed Implementation
[0064] Embodiments of this disclosure will now be described in more detail with reference to the accompanying drawings. While some embodiments of this disclosure are shown in the drawings, it should be understood that this disclosure can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of this disclosure. It should be understood that the accompanying drawings and embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of protection of this disclosure.
[0065] It should be understood that the steps described in the method embodiments of this disclosure may be performed in different orders and / or in parallel. Furthermore, the method embodiments may include additional steps and / or omit the steps shown. The scope of this disclosure is not limited in this respect.
[0066] The term "comprising" and its variations as used herein are open-ended, meaning "including but not limited to". The term "based on" means "at least partially based on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". Definitions of other terms will be given in the description below. It should be noted that the concepts of "first", "second", etc., used in this disclosure are only used to distinguish different devices, modules, or units, and are not intended to limit the order of functions performed by these devices, modules, or units or their interdependencies.
[0067] It should be noted that the terms "a" and "a plurality of" used in this disclosure are illustrative rather than restrictive, and those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".
[0068] The names of messages or information exchanged between multiple devices in the embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of such messages or information.
[0069] It is understood that before using the technical solutions disclosed in the various embodiments of this disclosure, users should be informed of the types, scope of use, and usage scenarios of the personal information involved in this disclosure in an appropriate manner in accordance with relevant laws and regulations, and user authorization should be obtained.
[0070] For example, upon receiving a user's active request, a prompt message is sent to the user to explicitly inform them that the requested operation will require the acquisition and use of the user's personal information. This allows the user to independently choose whether to provide personal information to the software or hardware, such as the electronic device, application, server, or storage medium performing the operations of this disclosed technical solution, based on the prompt message.
[0071] As an optional but non-limiting implementation, in response to a user's active request, sending a prompt message to the user can be done via a pop-up window, where the prompt message can be presented in text format. Furthermore, the pop-up window can also include a selection control allowing the user to choose "agree" or "disagree" to provide personal information to the electronic device. It is understood that the above notification and user authorization process is merely illustrative and does not constitute a limitation on the implementation of this disclosure; other methods that comply with relevant laws and regulations may also be applied to the implementation of this disclosure.
[0072] In one embodiment, such as Figure 1As shown, a system for detecting thermal stress in an aero-engine blade is provided, comprising: a green laser 2, a near-infrared laser 1, and a beam splitter 3, a scanning control unit 9, a signal receiver 10, and a processing module 8 connected in sequence. The green laser 2 and the near-infrared laser 1 are respectively connected to the beam splitter 3. The green laser 2 is used to output a first laser beam to the engine blade, and the near-infrared laser 1 is used to output a second laser beam to the engine blade. The beam splitter 3 is used to merge the first laser beam and the second laser beam to obtain a composite beam. The scanning control unit 9 is used to control the composite beam to scan the engine blade. The signal receiver 10 is used to receive ultrasonic signals and photoacoustic signals generated by the engine blade based on the composite beam. The processing module 8 is used to process the ultrasonic signals and photoacoustic signals to generate detection results. The detection results include at least: the maximum stress value of the engine blade, crack coordinates, and safety factor.
[0073] In one embodiment, such as Figure 2 As shown, the scanning control unit 9 includes a two-dimensional galvanometer 4 and a six-axis robotic arm 5. The two-dimensional galvanometer 4 is used to reflect the composite beam to generate a reflected beam, and the six-axis robotic arm 5 is used to drive the two-dimensional galvanometer 4 to move so that the reflected beam scans the engine blades according to a preset scanning trajectory. The signal receiver 10 includes a laser interferometer 7 and an infrared thermal imager 6. The laser interferometer 7 is used to receive the ultrasonic signal generated by the engine blades based on the first laser beam, and the infrared thermal imager 6 is used to receive the photoacoustic signal generated by the engine blades based on the second laser beam.
[0074] In this embodiment, the operator can control the simultaneous activation of the green laser 2 and the near-infrared laser 1. The beam splitter combines the first laser beam emitted by the green laser 2 and the second laser beam emitted by the near-infrared laser 1 into a single composite beam. After being reflected by the two-dimensional galvanometer 4, the beam is scanned on the surface of the engine blades by the six-axis robotic arm 5. The green laser 2 excites the engine blades to generate ultrasonic signals, which are received by the laser interferometer 7. The near-infrared laser 1 modulates the material stress of the engine blades, and the resulting photoacoustic signal is received by the infrared thermal imager 6. The processing module 8 processes the acoustic and photoacoustic signals to generate detection results.
[0075] In one possible embodiment, the wavelength of the green laser 2 can be 532 nm. This wavelength has high photon energy, making it easy to generate an effective photoacoustic effect with the surface of metallic materials, and can efficiently excite ultrasonic surface waves for high-sensitivity surface residual stress detection. The wavelength of the near-infrared laser 1 can be 1064 nm. This wavelength has strong penetrating power and is suitable for controlling the thermal stress inside the engine blade material and exciting deep photoacoustic signals, facilitating the detection of internal stress in thick-walled or complex structures. It should be noted that the above embodiments are merely illustrative of the wavelength selection of the near-infrared laser 1 and the green laser 2. In practical applications, the wavelengths of the near-infrared laser 1 and the green laser 2 can be flexibly adjusted according to the material of the engine blade, and are not limited here.
[0076] This disclosure provides a system and method for detecting thermal stress in an aero-engine blade. The system includes: a green laser, a near-infrared laser, and a beam splitter, a scanning control unit, a signal receiver, and a processing module connected in sequence. The green laser and the near-infrared laser are respectively connected to the beam splitter. The green laser outputs a first laser beam to the engine blade. The near-infrared laser outputs a second laser beam to the engine blade. The beam splitter merges the first and second laser beams to obtain a composite beam. The scanning control unit controls the composite beam to scan the engine blade. The signal receiver receives ultrasonic and photoacoustic signals generated by the engine blade based on the composite beam. The processing module processes the ultrasonic and photoacoustic signals to generate detection results. The detection results include at least: the maximum stress value of the engine blade, crack coordinates, and a safety factor.
[0077] As described above, the aero-engine blade thermal stress detection system provided by this disclosure uses a green laser and a near-infrared laser to generate a composite beam through a beam splitter. This composite beam, controlled by a scanning control unit, can achieve precise scanning of complex curved surfaces. Simultaneously, a signal receiver captures ultrasonic and photoacoustic signals, and a processing module further analyzes and processes these signals to generate detection results including the maximum stress value, crack coordinates, and safety factor of the engine blade. This aero-engine blade thermal stress detection system overcomes the limitations of single-mode detection in traditional technologies, achieving simultaneous acquisition of ultrasonic and photoacoustic signals. This allows for detection depth coverage of the entire blade thickness, avoiding the shortcomings of traditional technologies that can only detect the surface layer. Furthermore, by controlling the composite beam to scan the engine blade through the scanning control unit, this system can adapt to the real-time monitoring needs of complex operating conditions such as aero-engine start-up, shutdown, and variable loads, thereby improving the lifespan of aero-engine blades and flight safety assurance capabilities.
[0078] In one embodiment, such as Figure 3As shown, this embodiment relates to a method for detecting thermal stress in aero-engine blades using a detection system for thermal stress in aero-engine blades in any of the above embodiments. The method includes the following steps:
[0079] Step 301: Control the green laser to output a first laser beam to the engine blades, and control the near-infrared laser to output a second laser beam to the engine blades.
[0080] In one possible embodiment, the operator can control the simultaneous activation of the green laser 2 and the near-infrared laser 1. The green laser 2 outputs a first laser beam to the engine blades, which can excite the engine blades to generate ultrasonic signals. The near-infrared laser 1 outputs a second laser beam to the engine blades, which can excite the engine blades to generate photoacoustic signals.
[0081] Step 302: Combine the first laser beam and the second laser beam to obtain a composite beam.
[0082] In one possible embodiment, the beam splitter 3 can combine the first laser beam output from the green laser 2 to the engine blades and the second laser beam output from the near-infrared laser 1 to the engine blades to obtain a composite beam.
[0083] Step 303: Control the synthetic beam to scan the engine blades.
[0084] In one possible embodiment, after the synthesized beam is reflected by the two-dimensional galvanometer 4, the operator can control the six-axis robotic arm 5 to drive the synthesized beam to scan the surface of the engine blades.
[0085] Step 304: Receive ultrasonic and photoacoustic signals generated by the engine blades based on the synthetic beam.
[0086] In one possible embodiment, the laser interferometer 7 receives ultrasonic signals generated by the engine blades based on a composite beam, and the infrared thermal imager 6 receives photoacoustic signals generated by the engine blades based on a composite beam.
[0087] Step 305: Process the ultrasonic and photoacoustic signals to generate detection results.
[0088] In one possible embodiment, the laser interferometer 7 and the infrared thermal imager 6 send the received ultrasonic signals and photoacoustic signals to the processing module 8, respectively. The processing module 8 processes the ultrasonic signals and photoacoustic signals to generate detection results, wherein the detection results include at least: the maximum stress value of the engine blade, the crack coordinates, and the safety factor.
[0089] In one possible embodiment, such as Figure 4As shown, the process of processing ultrasonic and photoacoustic signals to generate detection results also includes the following steps:
[0090] Step 3051: Obtain the implicit field characteristics of the engine blades based on ultrasonic and photoacoustic signals.
[0091] Here, the processing module can acquire the implicit field characteristics of the engine blade based on ultrasonic and photoacoustic signals. These implicit field characteristics are used to characterize the physical state of stress, temperature, and material response at each point inside the engine blade.
[0092] In one possible embodiment, the implicit field characteristics of the engine blades are obtained based on ultrasonic and photoacoustic signals, including the following steps:
[0093] Determining the propagation delay of ultrasonic waves based on ultrasonic signals;
[0094] Calculate the elastic modulus and strain rate sensitivity coefficient of the engine blades;
[0095] The temperature gradient of the engine blades is obtained based on photoacoustic signals, and the coefficient of thermal expansion is calculated based on the temperature gradient.
[0096] The ultrasonic propagation delay, temperature gradient, elastic modulus, coefficient of thermal expansion, and strain rate sensitivity coefficient are input into the dynamic hash coding network model to obtain the implicit neural radiation field.
[0097] The implicit field characteristics of engine blades were determined based on the implicit neural radiation field.
[0098] Specifically, firstly, the processing module determines the ultrasonic propagation delay based on the ultrasonic signal. Then, it calculates the elastic modulus and strain rate sensitivity coefficient of the engine blade. Next, the processing module obtains the temperature gradient of the engine blade based on the photoacoustic signal and calculates the thermal expansion coefficient based on the temperature gradient. Then, the processing module inputs the ultrasonic propagation delay, temperature gradient, elastic modulus, thermal expansion coefficient, and strain rate sensitivity coefficient into the dynamic hash coding network model to obtain the implicit neural radiation field. The implicit neural radiation field is used to characterize the spatiotemporal evolution of the multi-physics coupling state of stress, temperature, and strain of the engine blade in three-dimensional space. Finally, the processing module determines the implicit field characteristics of the engine blade based on the implicit neural radiation field.
[0099] In one possible embodiment, near-infrared laser 1 and green laser 2 excite the engine blades to generate ultrasonic signals and temperature field changes. Laser interferometer 7 detects and records the ultrasonic response signal, which is the ultrasonic signal. Processing module 8 determines the ultrasonic propagation delay t using the ultrasonic signal and calculates the elastic modulus C of the engine blades. ijkl(T) and the strain rate sensitivity coefficient β, the infrared thermal imager 6 captures the temperature distribution of the engine blades in real time to obtain photoacoustic signals. The processing module 8 obtains the temperature gradient ▽T of the engine blades through the photoacoustic signals, and calculates the thermal expansion coefficient α based on the temperature gradient ▽T. T .
[0100] Processing module 8 obtains the ultrasonic wave propagation delay t, temperature gradient ▽T, and elastic modulus C. ijkl (T), coefficient of thermal expansion α T After the strain rate sensitivity coefficient β, t, C ijkl (T), α T β, as the basis for characterizing the local material properties and spatiotemporal evolution of engine blades, is input into the dynamic hash coding network model to obtain the implicit neural radiation field. It should be noted that the input of the traditional neural radiation field algorithm is the geometric position and viewpoint direction. The spatial radiation field is implicitly modeled through a multilayer perceptron (MLP), and the calculation formula is as follows:
[0101]
[0102] Where γ(.) is the hash-encoded output value, σ is the volume density, and c is the color feature. It can be seen that traditional algorithms rely solely on geometric position and viewpoint direction, failing to integrate multi-physics parameters and lacking physical conservation constraints. This embodiment makes the following improvement: it uses dynamic hash encoding to output the implicit neural radiation field, calculated as follows:
[0103]
[0104] Where H(x,y,z,t) is the hash-encoded output value, MLP is the neural network function, and (x,y,z,t) and l d For dynamic hierarchical parameters, Temperature gradient, λ s M is the feature space size and is the precision control parameter. This implicit neural radiation field is used to characterize the spatiotemporal evolution of the stress, temperature and strain multi-physics field coupling state of engine blades in three-dimensional space. It should be noted that in this embodiment, acoustic parameters, i.e. ultrasonic excitation characteristics, and thermodynamic parameters, i.e. temperature gradient, are jointly encoded as implicit field inputs, which breaks through the limitation of traditional neural radiation field algorithms that only rely on geometric position.
[0105] After obtaining the implicit neural radiation field, processing module 8 determines the implicit field characteristics of the engine blades based on the implicit neural radiation field. Processing module 8 concatenates the sound velocity tensor with the temperature gradient and inputs it into the MLP to generate implicit field characteristics. The calculation formula is as follows:
[0106]
[0107] in, The output of the implicit neural radiation field, i.e., the implicit field characteristics, C ijkl (x,y,z,t) is the elastic modulus. It should be noted that, through implicit neural field mapping, the processing module 8 fuses the acoustic and thermal information to generate implicit field features, providing an efficient and continuous spatial feature representation for the subsequent differential rendering field, thereby realizing a physically driven and smooth three-dimensional stress field reconstruction representation.
[0108] Step 3052: Determine the stress tensor of the engine blades based on implicit field characteristics.
[0109] Here, after the processing module 8 obtains the implicit field characteristics of the engine blade based on ultrasonic and photoacoustic signals, it can determine the stress tensor of the engine blade based on the implicit field characteristics. The stress tensor is used to characterize the magnitude, direction and thermo-mechanical coupling state of the stress borne by any point inside the engine blade in three-dimensional space.
[0110] In one possible embodiment, determining the stress tensor of the engine blades based on implicit field characteristics includes the following steps:
[0111] The implicit field features are input into the multilayer perceptron network, and physical constraint equations are embedded in the multilayer perceptron network to obtain the modified multilayer perceptron network.
[0112] The stress tensor of the engine blades is determined based on the modified multilayer perceptron network.
[0113] Specifically, firstly, the processing module 8 inputs the implicit field features into the multilayer perceptron network and embeds physical constraint equations into the multilayer perceptron network to obtain a modified multilayer perceptron network. The physical constraint equations include at least the momentum conservation equation and the energy conservation equation. Then, the processing module 8 determines the stress tensor of the engine blade based on the modified multilayer perceptron network.
[0114] It should be noted that traditional data-driven models are usually based on local data fitting and ignore multi-field coupling effects. The traditional formula for calculating the stress tensor of engine blades is:
[0115]
[0116] Where, ω i (x,y,z) is a weight function based on local features, f i (t) is the time-series stress response function. According to the above formula, it can be observed that although the traditional data-driven model can quickly predict stress through simple mathematical expressions, it is limited by single-factor input and static constitutive assumptions, and lacks explicit modeling of mechanical and thermal conservation laws, making it difficult to apply to the actual working conditions of aero-engine blades.
[0117] In this embodiment, the processing module 8 inputs the implicit field features into the multilayer perceptron network and embeds physical constraint equations into the multilayer perceptron network to obtain a modified multilayer perceptron network. The physical constraint equations include at least the momentum conservation equation and the energy conservation equation. Specifically, the processing module 8 uses the elastic modulus C... ijkl (T), coefficient of thermal expansion α kl The constitutive relation is constructed using the strain rate sensitivity coefficient β, and dual physical constraints of momentum conservation and energy conservation are applied. The formula for the modified multilayer perceptron network is as follows:
[0118]
[0119] Where, σ ij Let δ be the stress tensor. ij For the Kronecker function, This represents the equivalent rate of change.
[0120] The momentum conservation equation embedded in a multilayer perceptron network is:
[0121]
[0122] Where ρ is the material density and u is the displacement field.
[0123] The energy conservation equation embedded in a multilayer perceptron network is:
[0124]
[0125] Among them, C p ρ is the specific heat capacity, k is the thermal conductivity. Let Q be the strain rate tensor and Q be the heat source.
[0126] It should be noted that in this embodiment, the processing module 8 explicitly integrates the material constitutive relation (stress-strain-temperature coupling) with the Navier-Stokes (momentum) and Fourier (energy) equations to realize the physical-driven multi-scale stress field evolution, so as to ensure the physical consistency between the numerical and neural field results and improve the reliability of the results.
[0127] Step 3053: Establish a three-dimensional stress field distribution cloud map of the engine blade based on the stress tensor, and determine the detection results based on the three-dimensional stress field distribution cloud map.
[0128] Here, the processing module 8 can establish a three-dimensional stress field distribution cloud map of the engine blade based on the stress tensor, and determine the detection results based on the three-dimensional stress field distribution cloud map.
[0129] In one possible embodiment, establishing a three-dimensional stress field distribution cloud map of the engine blade based on the stress tensor includes the following steps:
[0130] The stress tensor is mapped to a continuous three-dimensional stress field function through differentiable interpolation;
[0131] The continuous three-dimensional stress field function is optimized by using a loss function to obtain the optimized three-dimensional stress field function.
[0132] A three-dimensional stress field distribution cloud map is established based on the optimized three-dimensional stress field function.
[0133] Specifically, firstly, the processing module 8 maps the stress tensor to a continuous three-dimensional stress field function through differentiable interpolation. Then, the processing module 8 optimizes the continuous three-dimensional stress field function through a loss function to obtain the optimized three-dimensional stress field function. Finally, the processing module 8 establishes a three-dimensional stress field distribution cloud map based on the optimized three-dimensional stress field function.
[0134] In one possible embodiment, processing module 8 maps the stress tensor to a continuous three-dimensional stress field function through differentiable interpolation. The differentiable interpolation formula is as follows:
[0135]
[0136] Where σ(p) is the stress value at point p, ω k The weight coefficients for key point k can be dynamically adjusted through hash encoding, p k Let ω be the coordinates of the discrete sampling points, σ be the weight vector, and σ be the weight vector. k The actual stress value at discrete sampling point k.
[0137] Processing module 8 optimizes the continuous three-dimensional stress field function using a loss function, where the loss function for the differentiable interpolation formula is:
[0138]
[0139] Among them, W i For each location error weighting factor, λ represents the actual stress value, and λ is the regularization parameter used to suppress weighted noise.
[0140] After obtaining the optimized three-dimensional stress field function, the processing module 8 establishes a three-dimensional stress field distribution cloud map based on the optimized three-dimensional stress field function, which can realize high-precision continuous differentiable stress field visualization rendering of engine blades based on physical constraints.
[0141] In one possible embodiment, determining the detection result based on a three-dimensional stress field distribution cloud map includes the following steps:
[0142] An interactive interface is used to dynamically render the three-dimensional stress field distribution cloud map to obtain a visual rendering result.
[0143] The detection results are determined based on the visual rendering results.
[0144] Specifically, after the processing module 8 establishes the three-dimensional stress field distribution cloud map, the processing module 8 first performs interactive dynamic rendering of the three-dimensional stress field distribution cloud map to obtain a visual rendering result. The visual rendering result is a color gradient display of the stress gradient of the engine blade. Then, the processing module 8 determines the detection result based on the visual rendering result. The detection result includes at least the maximum stress value of the engine blade, crack coordinates, and safety factor.
[0145] In one possible embodiment, the processing module 8 can dynamically render the three-dimensional stress field distribution cloud map constructed in the above embodiment through an interactive 3D interface, combine it with OpenGL technology to realize the color gradient display of stress gradient, and support rotation, scaling and translation operations.
[0146] In one possible embodiment, processing module 8 can extract the isosurfaces corresponding to the stress threshold on the three-dimensional stress field distribution cloud map using the MarchingCubes algorithm, intuitively presenting the shape and distribution range of the stress concentration area. Processing module 8 can automatically mark crack hotspots using LSTM prediction residual data, indicating the propagation direction with three-dimensional arrows; and can simultaneously overlay infrared thermal imaging temperature field and laser interferometric strain field data to achieve multi-physics spatiotemporal coupling analysis. Through the above processing, the final processing module 8 can generate a test report that conforms to national standards, which includes key parameters such as the maximum stress value of the engine blade, crack coordinates, and safety factor.
[0147] In this embodiment, firstly, the processing module acquires the implicit field characteristics of the engine blade based on the ultrasonic signal and the photoacoustic signal. Then, the processing module determines the stress tensor of the engine blade based on the implicit field characteristics. Finally, the processing module establishes a three-dimensional stress field distribution cloud map of the engine blade based on the stress tensor and determines the detection result based on the three-dimensional stress field distribution cloud map.
[0148] As described above, this embodiment sequentially acquires implicit field features, determines the stress tensor, establishes a three-dimensional stress field distribution cloud map, and generates detection results through the processing module. This achieves deep fusion of multiple physical field parameters and high-precision stress characterization under physical constraints. Specifically, this embodiment utilizes parameters such as propagation delay and temperature gradient extracted from ultrasonic and photoacoustic signals to construct a dynamic hash coding network model, breaking through the limitation of traditional neural radiation fields relying solely on geometric positions and integrating acoustic and thermodynamic parameters into implicit field features. This embodiment embeds momentum and energy conservation equations into a multilayer perceptron network to achieve strong coupling modeling of stress-strain-temperature fields, solving the problem of traditional data-driven models lacking physical conservation. This embodiment uses differentiable interpolation and loss function optimization to map the stress tensor into a continuous three-dimensional stress field function. Combined with interactive 3D rendering and multi-physics field superposition analysis, it intuitively presents stress concentration areas and crack hotspots, ultimately generating a detection report containing key parameters such as maximum stress value and crack coordinates. This provides a solution for thermal stress detection of aero-engine blades that combines physical accuracy and visualization efficiency, effectively improving the comprehensiveness and reliability of stress assessment under complex working conditions.
[0149] In one possible embodiment, after establishing a three-dimensional stress field distribution cloud map of the engine blade based on the stress tensor, the following steps are also included:
[0150] The local stress intensity of engine blades is determined based on the three-dimensional stress field distribution cloud map;
[0151] The updated hash coding network model is obtained by updating the dynamic hash resolution of the dynamic hash coding network model based on the local stress intensity.
[0152] The updated three-dimensional stress field distribution cloud map is determined based on the updated hash-coded network model.
[0153] Specifically, after establishing a three-dimensional stress field distribution cloud map of the engine blade based on the stress tensor, processing module 8 determines the local stress intensity of the engine blade based on the three-dimensional stress field distribution cloud map, and updates the dynamic hash resolution of the dynamic hash coding network model based on the local stress intensity. That is, the updated hash coding network model is obtained through the hash resolution adaptive algorithm. The calculation process of hash resolution adaptation is as follows:
[0154]
[0155] Among them, L h L0 is the dynamic hash resolution, β is the initial resolution, and σ is the stress sensitivity coefficient. vM For VM equivalent stress, σ refTo reference the baseline stress, after obtaining the updated hash-coded network model, the processing module 8 can determine the updated three-dimensional stress field distribution cloud map based on the updated hash-coded network model.
[0156] As described above, in this embodiment, after establishing a three-dimensional stress field distribution cloud map based on the stress tensor, the processing module dynamically updates the resolution of the hash coding network model by determining the local stress intensity, thereby achieving adaptive optimization of detection accuracy. Specifically, the local stress intensity calculated by the processing module based on the three-dimensional stress field distribution cloud map can drive the nonlinear adjustment of the dynamic hash resolution, enabling the system to automatically increase the sampling density in stress concentration areas to accurately capture stress peak details, while reducing the resolution in stress flat areas to balance the computational load. This adaptive mechanism breaks through the limitations of traditional fixed resolution detection, ensuring the detection accuracy of key areas and improving the overall computational efficiency. This allows the system to more efficiently cope with the dynamic changes in the stress field under extreme conditions such as the start-up and shutdown of aero-engine blades and variable loads, providing a more intelligent technical solution for the accurate assessment of blade thermal stress.
[0157] In one embodiment, such as Figure 5 As shown, the method for detecting thermal stress in aero-engine blades also includes the following steps:
[0158] Step 501: The reference stress, current stress field, strain rate and process random noise of the engine blade are processed by a long short-term memory network to obtain the stress field of the engine blade at the next moment.
[0159] In one possible embodiment, processing module 8 processes the reference stress of the engine blade, the stress field at the current moment, the strain rate, and the process random noise through a long short-term memory network to obtain the stress field of the engine blade at the next moment. The calculation formula is as follows:
[0160]
[0161] Where, σ t+1 The stress field at the next time step, LSTM is a long short-term memory network, σ t For the stress field at the current moment, strain rate, n t This represents random noise in the process.
[0162] Step 502: Determine the gradient distribution of the engine blades based on the stress field at the next moment.
[0163] In one possible embodiment, processing module 8 obtains the stress field σ of the engine blade at the next moment. t+1 Then, the gradient distribution of thermal stress on the engine blades can be determined.
[0164] Step 503: Determine the scanning command of the six-axis robotic arm based on the gradient distribution.
[0165] In one possible embodiment, the processing module 8 can determine the scanning command of the six-axis robotic arm 5 based on the gradient distribution. The scanning command is used to drive the six-axis robotic arm 5 to move the two-dimensional galvanometer 4, so that the reflected light beam can perform a high-density scan on the high-stress gradient region of the engine blade.
[0166] As described above, in this embodiment, the processing module 8 uses a Long Short-Term Memory (LSTM) network to perform time-series processing on the reference stress, current stress field, strain rate, and process random noise, accurately predicts the stress field at the next moment, and then dynamically generates six-axis robotic arm scanning instructions based on the stress gradient distribution, realizing intelligent optimization of the scanning path and further improving the reliability of the detection method for thermal stress of aero-engine blades.
[0167] In one embodiment, such as Figure 6 As shown, by combining all the above embodiments, an overall process for a method of detecting thermal stress in aero-engine blades is provided.
[0168] In this embodiment, the overall process of the method for detecting thermal stress in aero-engine blades includes seven stages: data input, nonlinear dynamic topology encoding, implicit neural field mapping, multi-physics field strong coupling modeling constrained by conservation laws, differentiable rendering, LSTM-Hash collaborative resolution control, and adaptive visualization and crack analysis of multimodal spatiotemporal coupled stress field.
[0169] The specific operations in the "data input" stage are as follows: the laser interferometer 7 and the infrared thermal imager 6 input the received ultrasonic and photoacoustic signals to the processing module 8, respectively; the specific operations in the "nonlinear dynamic topology coding" stage are as follows: the processing module 8 performs topology coding on the input ultrasonic and photoacoustic signals using nonlinear dynamic system theory. By analyzing the dynamic correlation between data such as ultrasonic propagation delay and temperature gradient, it extracts nonlinear dynamic topology features such as stress-strain and temperature-deformation inside the blade, that is, it constructs a phase space model of the data, mines the topological structure of different physical quantities changing with time and space, and transforms the original signal into coded features containing dynamic correlation information; the specific operations in the "implicit neural field mapping" stage are as follows. To build an implicit neural radiation field (NeRF) model for processing module 8, the features encoded by nonlinear dynamic topology are used as input and mapped to an implicit space. The model learns the implicit representations of physical fields such as blade stress, temperature, and strain in three-dimensional space through a multilayer perceptron (MLP), achieving unified modeling of multiple physical quantities. The specific operation of the "conservation law-constrained multi-physics field strong coupling modeling" stage involves processing module 8 embedding physical conservation law constraints on the implicit neural field mapping, introducing the momentum conservation equation, and combining it with the constitutive relation of the blade material to correct the implicit neural field results, ensuring that the modeling conforms to real physical laws and achieving strong coupling of multiple physics fields. The specific operation of the "differentiable rendering" stage involves processing module 8 building upon the strong coupling model... The implicit neural field after modeling employs a differentiable rendering algorithm to transform the abstract physical field into an intuitive visualization result. Processing module 8, by constructing a differentiable rendering pipeline, maps physical quantities such as stress field and temperature field into visualization forms such as 3D cloud maps and isosurfaces, while preserving differential properties and supporting subsequent backpropagation optimization. The specific operation of the "LSTM-Hash collaborative resolution control" stage involves processing module 8 using a Long Short-Term Memory (LSTM) network to learn the temporal evolution law of the blade stress field, predicting the stress gradient distribution at different times, and constructing an efficient spatial index using a hash algorithm. Based on the stress gradient distribution predicted by LSTM, the resolution of differentiable rendering and subsequent analysis is dynamically adjusted, focusing on stress concentration, ... In areas with large gradients (such as the leading edge of the blade and the tenon), the resolution is increased to accurately capture details, while the resolution is reduced in areas with gentle stress to balance computational efficiency and accuracy. The specific operation of the "Adaptive Visualization and Crack Analysis of Multimodal Spatiotemporal Coupled Stress Field" stage is to process the stress field visualization results after adaptive resolution adjustment by module 8, perform crack identification and evaluation, use the MarchingCubes algorithm to extract stress threshold isosurfaces to present the morphology of stress concentration areas, use LSTM to predict residual markers of crack hotspots and indicate the direction of propagation, and superimpose multiphysics data to carry out spatiotemporal coupling analysis, and finally generate an inspection report, which includes key parameters such as maximum stress value, crack coordinates, and safety factor.
[0170] In one possible embodiment, to verify the superiority of the thermal stress detection system for aero-engine blades in the above embodiment, thermal stress detection was performed on TC4 titanium alloy aero-engine blades through experiments. The experimental results show that the thermal stress detection system for aero-engine blades in the above embodiment achieves the technical specifications of spatial resolution of 0.3 mm and absolute error of thermal stress detection of ±18 MPa in the detection of TC4 titanium alloy aero-engine blades, with a crack detection rate of 95%. In the blade root region with a thickness of 3.2 mm, an internal stress gradient of up to 18 MPa / mm was detected. However, traditional surface detection can only reflect a surface gradient of ≤5 MPa / mm.
[0171] The above experiments demonstrate that the method for detecting thermal stress in aero-engine blades provided in the above embodiments achieves multi-dimensional technological breakthroughs compared to traditional X-ray diffraction (XRD): In terms of detection depth, it can penetrate the entire thickness, overcoming the limitation of XRD being limited to the surface layer of 0.05-0.1 mm, improving the spatial resolution to 0.3 mm, and establishing a stress gradient-driven adaptive mechanism; in terms of dynamic detection, it supports real-time tracking of transient thermal stress, achieving a leap in magnitude compared to the static measurement of XRD, while breaking through the limitations of destructive sample preparation, achieving in-situ non-destructive testing through multi-modal laser collaboration, and integrating differential-algebraic dual conservation constraints to construct a three-dimensional volumetric stress field distribution model, completely solving the distortion problem of traditional two-dimensional surface stress characterization. Finally, in terms of operating condition adaptability, it extends from the laboratory to extreme industrial environments, with a detection speed of 200 mm² / s, meeting the online monitoring requirements of aero-engines under all operating conditions.
[0172] As described above, this disclosed technical solution achieves a systematic breakthrough in the detection of thermal stress in aero-engine blades by innovatively integrating multimodal laser detection and dynamic neural radiation field algorithms. The solution employs nonlinear dynamic topology coding to construct an implicit radiation field, and for the first time, jointly models acoustoelastic parameters and temperature gradients, overcoming the limitations of traditional algorithms that rely on geometric location. Through a conservation law-constrained multi-physics field strongly coupled model, this solution achieves strict physical consistency in the evolution of stress-strain-temperature fields, significantly improving the reliability of the results. Combined with an LSTM-Hash collaborative resolution control mechanism, this solution dynamically optimizes the scanning path and adaptively improves the accuracy of stress concentration areas, meeting the real-time monitoring requirements under extreme operating conditions. Finally, this solution constructs a penetrating three-dimensional dynamic stress field visualization system, supporting sub-micron level precision detection and internal stress gradient analysis of complex surfaces, simultaneously achieving crack hotspot location and multi-physics field spatiotemporal coupling diagnosis. This provides a high-precision, non-contact, and fully adaptable intelligent non-destructive testing solution for the prediction of lifespan of key aero-engine components and for ensuring flight safety.
[0173] This disclosure also provides an electronic device, including: at least one processor; a memory for storing processor-executable instructions; wherein the at least one processor is configured to execute the instructions to implement the methods disclosed in this disclosure.
[0174] Figure 7 This is a schematic diagram of the structure of an electronic device provided as an exemplary embodiment of this disclosure. For example... Figure 7 As shown, the electronic device 700 includes at least one processor 701 and a memory 702 coupled to the processor 701. The processor 701 can perform the corresponding steps in the methods disclosed in the embodiments of this disclosure.
[0175] The processor 701 described above can also be called a central processing unit (CPU), which can be an integrated circuit chip with signal processing capabilities. Each step in the method disclosed in this embodiment can be implemented by the integrated logic circuitry in the processor 701 hardware or by software instructions. The processor 701 can be a general-purpose processor, a digital signal processor (DSP), an ASIC, a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this embodiment can be directly implemented by a hardware decoding processor, or implemented by a combination of hardware and software modules in the decoding processor. The software modules can be located in the memory 702, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. The processor 701 reads information from the memory 702 and, in conjunction with its hardware, completes the steps of the method described above.
[0176] Furthermore, various operations / processes according to this disclosure, implemented via software and / or firmware, can be transmitted from a storage medium or network to a computer system with a dedicated hardware architecture, such as... Figure 8 The computer system 800 shown is equipped with the programs that constitute the software. When various programs are installed, the computer system is able to perform various functions, including those described above. Figure 8 A block diagram of a computer system provided for an exemplary embodiment of this disclosure.
[0177] Computer system 800 is intended to represent various forms of digital electronic computer devices, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. Electronic devices may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present disclosure described and / or claimed herein.
[0178] like Figure 8 As shown, the computer system 800 includes a computing unit 801, which can perform various appropriate actions and processes based on a computer program stored in a read-only memory (ROM) 802 or a computer program loaded from a storage unit 808 into a random access memory (RAM) 803. The RAM 803 may also store various programs and data required for the operation of the computer system 800. The computing unit 801, ROM 802, and RAM 803 are interconnected via a bus 804. An input / output (I / O) interface 805 is also connected to the bus 804.
[0179] Multiple components in the computer system 800 are connected to the I / O interface 805, including: an input unit 806, an output unit 807, a storage unit 808, and a communication unit 809. The input unit 806 can be any type of device capable of inputting information into the computer system 800. The input unit 806 can receive input numerical or character information and generate key signal inputs related to user settings and / or function control of the electronic device. The output unit 807 can be any type of device capable of presenting information and may include, but is not limited to, a monitor, speaker, video / audio output terminal, vibrator, and / or printer. The storage unit 808 may include, but is not limited to, a hard disk and an optical disk. The communication unit 809 allows the computer system 800 to exchange information / data with other devices via a network such as the Internet, and may include, but is not limited to, a modem, network card, infrared communication device, wireless communication transceiver, and / or chipset, such as Bluetooth™ device, WiFi device, WiMax device, cellular communication device, and / or the like.
[0180] The computing unit 801 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 801 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 801 performs the various methods and processes described above. For example, in some embodiments, the methods disclosed in this disclosure can be implemented as a computer software program tangibly contained in a machine-readable medium, such as storage unit 808. In some embodiments, part or all of the computer program can be loaded and / or installed on the electronic device 700 via ROM 802 and / or communication unit 809. In some embodiments, the computing unit 801 can be configured to perform the methods disclosed in this disclosure by any other suitable means (e.g., by means of firmware).
[0181] This disclosure also provides a computer-readable storage medium, wherein when the instructions in the computer-readable storage medium are executed by a processor of an electronic device, the electronic device is able to perform the methods disclosed in this disclosure.
[0182] The computer-readable storage medium in this disclosure can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. The aforementioned computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specifically, the aforementioned computer-readable storage medium may include electrical connections based on one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.
[0183] The aforementioned computer-readable medium may be included in the aforementioned electronic device; or it may exist independently and not assembled into the electronic device.
[0184] Figure 9 A computer program product 900 is provided as an exemplary embodiment of the present disclosure. The computer program product 900 includes a computer program 901, wherein the computer program 901, when executed by a processor, implements the methods disclosed in the embodiments of the present disclosure.
[0185] In embodiments of this disclosure, computer program code for performing the operations of this disclosure can be written in one or more programming languages or a combination thereof. These programming languages include, but are not limited to, object-oriented programming languages such as Java, Smalltalk, and C++, as well as conventional procedural programming languages such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network (including a local area network (LAN) or a wide area network (WAN)), or it can be connected to an external computer.
[0186] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.
[0187] The modules, components, or units described in the embodiments of this disclosure can be implemented in software or hardware. The names of the modules, components, or units do not necessarily constitute a limitation on the module, component, or unit itself.
[0188] The functions described above in this document can be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary hardware logic components that can be used include: field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), system-on-a-chip (SoCs), complex programmable logic devices (CPLDs), and so on.
[0189] The above description is merely an embodiment of this disclosure and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of this disclosure is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-described concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features disclosed in this disclosure that have similar functions.
[0190] While specific embodiments of this disclosure have been described in detail by way of example, those skilled in the art should understand that the examples are for illustrative purposes only and not intended to limit the scope of this disclosure. Those skilled in the art should understand that modifications can be made to the above embodiments without departing from the scope and spirit of this disclosure. The scope of this disclosure is defined by the appended claims.
Claims
1. A system for detecting thermal stress in aero-engine blades, characterized in that, The system includes: a green laser, a near-infrared laser, and a beam splitter, a scanning control unit, a signal receiver, and a processing module connected in sequence; the green laser and the near-infrared laser are respectively connected to the beam splitter; The green laser is used to output a first laser beam to the engine blades; The near-infrared laser is used to output a second laser beam to the engine blades; The beam splitter is used to combine the first laser beam and the second laser beam to obtain a composite beam; The scanning control unit is used to control the synthetic beam to scan the engine blades; The signal receiver is used to receive ultrasonic signals and photoacoustic signals generated by the engine blades based on the synthetic beam; The processing module is used to process the ultrasonic signal and the photoacoustic signal to generate a detection result; wherein the detection result includes at least: the maximum stress value of the engine blade, the crack coordinates, and the safety factor; The processing module is used for: The implicit field characteristics of the engine blade are obtained based on the ultrasonic signal and the photoacoustic signal; wherein, the implicit field characteristics are used to characterize the physical state of stress, temperature and material response at each point inside the engine blade; The stress tensor of the engine blade is determined based on the implicit field characteristics; wherein, the stress tensor is used to characterize the magnitude, direction and thermo-mechanical coupling state of the stress borne by any point inside the engine blade in three-dimensional space; A three-dimensional stress field distribution cloud map of the engine blade is established based on the stress tensor, and the detection result is determined based on the three-dimensional stress field distribution cloud map.
2. The system according to claim 1, characterized in that, The scanning control unit includes: a two-dimensional galvanometer and a six-axis robotic arm; The two-dimensional galvanometer is used to reflect the synthesized beam to generate the reflected beam; The six-axis robotic arm is used to drive the two-dimensional galvanometer to move, so that the reflected light beam scans the engine blades according to a preset scanning trajectory.
3. A method for detecting thermal stress in aero-engine blades, characterized in that, The method of the detection system for thermal stress on aero-engine blades according to any one of claims 1 to 2 includes: Control the green laser to output a first laser beam to the engine blades, and control the near-infrared laser to output a second laser beam to the engine blades; The first laser beam and the second laser beam are combined to obtain a composite beam; Control the composite beam to scan the engine blades; Receive ultrasonic and photoacoustic signals generated by the engine blades based on the synthetic beam; The ultrasonic signal and the photoacoustic signal are processed to generate a detection result; wherein the detection result includes at least: the maximum stress value of the engine blade, the crack coordinates, and the safety factor.
4. The method according to claim 3, characterized in that, The step of processing the ultrasonic signal and the photoacoustic signal to generate a detection result includes: The implicit field characteristics of the engine blade are obtained based on the ultrasonic signal and the photoacoustic signal; wherein, the implicit field characteristics are used to characterize the physical state of stress, temperature and material response at each point inside the engine blade; The stress tensor of the engine blade is determined based on the implicit field characteristics; wherein, the stress tensor is used to characterize the magnitude, direction and thermo-mechanical coupling state of the stress borne by any point inside the engine blade in three-dimensional space; A three-dimensional stress field distribution cloud map of the engine blade is established based on the stress tensor, and the detection result is determined based on the three-dimensional stress field distribution cloud map.
5. The method according to claim 4, characterized in that, The process of obtaining the implicit field characteristics of the engine blades based on the ultrasonic signal and the photoacoustic signal includes: The ultrasonic propagation delay is determined based on the ultrasonic signal; Calculate the elastic modulus and strain rate sensitivity coefficient of the engine blade; The temperature gradient of the engine blades is obtained based on the photoacoustic signal, and the coefficient of thermal expansion is calculated based on the temperature gradient. The ultrasonic wave propagation delay, temperature gradient, elastic modulus, thermal expansion coefficient, and strain rate sensitivity coefficient are input into a dynamic hash coding network model to obtain an implicit neural radiation field; wherein, the implicit neural radiation field is used to characterize the spatiotemporal evolution of the stress, temperature, and strain multiphysics coupling state of the engine blade in three-dimensional space. The implicit field characteristics of the engine blades are determined based on the implicit neural radiation field.
6. The method according to claim 4, characterized in that, The determination of the stress tensor of the engine blade based on the implicit field characteristics includes: The implicit field features are input into a multilayer perceptron network, and physical constraint equations are embedded in the multilayer perceptron network to obtain a modified multilayer perceptron network; wherein, the physical constraint equations include at least: momentum conservation equation and energy conservation equation; The stress tensor of the engine blade is determined based on the modified multilayer perceptron network.
7. The method according to claim 4, characterized in that, The process of establishing a three-dimensional stress field distribution cloud map of the engine blade based on the stress tensor includes: The stress tensor is mapped to a continuous three-dimensional stress field function by differentiable interpolation; The continuous three-dimensional stress field function is optimized by using a loss function to obtain the optimized three-dimensional stress field function. A three-dimensional stress field distribution cloud map is established based on the optimized three-dimensional stress field function.
8. The method according to claim 4, characterized in that, The determination of the detection result based on the three-dimensional stress field distribution cloud map includes: The three-dimensional stress field distribution cloud map is dynamically rendered through an interactive interface to obtain a visual rendering result; wherein, the visual rendering result is a color gradient display of the stress gradient of the engine blade; The detection results are determined based on the visualization rendering results; wherein, the detection results include at least: the maximum stress value of the engine blade, the crack coordinates, and the safety factor.
9. The method according to claim 4, characterized in that, After establishing a three-dimensional stress field distribution cloud map of the engine blade based on the stress tensor, the method further includes: The local stress intensity of the engine blades is determined based on the three-dimensional stress field distribution cloud map. The dynamic hash resolution of the dynamic hash coding network model is updated based on the local stress intensity to obtain the updated hash coding network model. The updated three-dimensional stress field distribution cloud map is determined based on the updated hash-encoded network model.
10. The method according to claim 4, characterized in that, The method further includes: The reference stress, current stress field, strain rate, and process random noise of the engine blade are processed by a long short-term memory network to obtain the stress field of the engine blade at the next moment. The gradient distribution of the engine blades is determined based on the stress field at the next moment; The scanning command for the six-axis robotic arm is determined based on the gradient distribution; wherein, the scanning command is used to drive the six-axis robotic arm to move the two-dimensional galvanometer, so that the reflected light beam performs a high-density scan on the high-stress gradient region of the engine blade.
Citation Information
Patent Citations
Pillar porcelain insulator stress monitoring device and method based on double laser scanning
CN113720508A
