Automatic production and intelligent detection method for aero-engine casing forgings

Abstract

The invention discloses an automatic production and intelligent detection method for aero-engine casing forgings, and particularly relates to the technical field of automatic production and intelligent detection. The thickness of the casing forgings is calculated according to ultrasonic propagation time, a thin-wall area is identified according to the thickness of the casing forgings, and medium staggering information of the thin-wall area is obtained; calculating a medium staggering coefficient according to the medium staggering information of the thin-wall area; the method comprises the following steps: acquiring recovery time information of an ultrasonic probe, and calculating a probe recovery time fluctuation coefficient according to the recovery time information of the ultrasonic probe; the method comprises the following steps: acquiring environment interference information of an ultrasonic signal, and calculating an environment interference coefficient according to the environment interference information of the ultrasonic signal; through quantitative analysis of the medium staggering degree of the thin-wall area, the probe recovery time fluctuation degree and the environment interference, the potential leak detection risk can be effectively identified, a multi-dimensional risk assessment model is adopted, and the detection efficiency and accuracy are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the field of automated production and intelligent detection technology, and more specifically, to an automated production and intelligent detection method for aircraft engine casing forgings. Background Art

[0002] The aircraft engine casing is an important part of the aircraft engine, and its design and manufacturing require extremely high precision and reliability. During the use of the aircraft engine, the casing is subjected to an extremely complex working environment, including high temperature, high pressure and severe mechanical stress, so the integrity of its structure is directly related to the safety and reliability of the engine. In order to ensure the quality and safety of the casing, accurate detection methods are particularly important, especially in the increasingly widespread application of automated production and intelligent detection.

[0003] Aircraft engine casings are usually forged from high-strength alloy materials and have complex geometric shapes, such as thin walls, inner cavities, and multiple holes. These characteristics pose great challenges to detection technology, especially in ultrasonic detection technology, where signal attenuation in thin-walled areas, near-surface blind areas, and edge effects seriously affect the accurate identification and positioning of defects.

[0004] To solve these problems, modern aircraft engine casing automation production and intelligent detection systems have begun to introduce more efficient and accurate technical means. Ultrasonic testing technology, as a non-destructive testing method, is widely used in the quality inspection of parts. However, in thin-walled areas, the ultrasonic signal propagation path is short and easy to attenuate. In addition, due to the way the probe contacts the surface and the interference of the signal, it may cause blind spots in the detection, missed detection or misjudgment of defects. Especially under the complex geometric shapes of forgings, traditional ultrasonic testing methods are difficult to cope with various changes in form, which further increases the complexity and uncertainty of detection. Summary of the invention

[0005] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides an automated production and intelligent detection method for aircraft engine casing forgings to solve the problems raised in the above-mentioned background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] The automated production and intelligent detection method of aircraft engine casing forgings includes the following steps:

[0008] Step S1, transmitting ultrasonic pulses to different positions of the casing forging, and using a receiver to capture echo signals to measure ultrasonic propagation time, calculating the thickness of the casing forging according to the ultrasonic propagation time, and identifying the thin-walled area according to the thickness of the casing forging;

[0009] Step S2, obtaining medium interleaving information of the thin-walled area and calculating a medium interleaving coefficient according to the medium interleaving information of the thin-walled area to evaluate the medium interleaving degree of the thin-walled area;

[0010] Step S3, obtaining recovery time information of the ultrasonic probe, and calculating the probe recovery time fluctuation coefficient according to the recovery time information of the ultrasonic probe, so as to evaluate the recovery time fluctuation degree of the ultrasonic probe;

[0011] Step S4, obtaining environmental interference information of the ultrasonic signal and calculating an environmental interference coefficient according to the environmental interference information of the ultrasonic signal to evaluate the influence of the environmental interference on the ultrasonic signal;

[0012] Step S5, constructing a casing forging missed inspection risk assessment model according to the medium interleaving coefficient, the probe recovery time fluctuation coefficient, and the environmental interference coefficient, outputting a casing forging missed inspection risk assessment index, and assessing the missed inspection risk of aircraft engine casing forgings.

[0013] In a preferred embodiment, in step S1, ultrasonic pulses are emitted to different positions of the casing forging, the ultrasonic pulse emission time is recorded, the echo signal is captured by a receiver and the echo signal capture time is recorded, and the ultrasonic propagation time is calculated according to the ultrasonic pulse emission time and the echo signal capture time, and the expression is as follows: csb = |hbt-fst|, wherein csb represents the ultrasonic propagation time, hbt represents the echo signal capture time, and fst represents the ultrasonic pulse emission time; the casing forging thickness is calculated according to the ultrasonic propagation time, and the expression is as follows: Where jxd represents the thickness of the casing forging, vs represents the sound velocity of the casing forging material;

[0014] The thickness of the casing forging is compared with a preset standard thickness, and the area where the thickness of the casing forging is less than the standard thickness is marked as a thin-walled area.

[0015] In a preferred embodiment, the medium interleaving degree of the thin-walled area is measured by acquiring the medium interleaving information of the thin-walled area, analyzing the medium interleaving degree of the thin-walled area, and calculating the medium interleaving coefficient;

[0016] The logic for obtaining the medium interleaving coefficient is as follows:

[0017] Get the echo signal x(t) from the thin-wall area and decompose the echo signal using wavelet transform, as follows: Where W ψ(a,b) represents the wavelet transform coefficient, a represents the scale factor, b represents the translation factor, ψ represents the mother wavelet function, and x(t) represents the echo signal at time t. The echo signal from the thin-walled area is divided into N time domain windows of the same interval, and the signal energy of each time domain window is calculated. The expression is as follows: E i =∑ a,b |W ψ (a,b)| 2 , where E i Represents the signal energy of the i-th time domain window; calculate the signal energy ratio, the expression is as follows: Where P i Represents the signal energy proportion of the i-th time domain window, i={1,2,...,N}, N is a positive integer; calculate the signal entropy value, the expression is as follows: Where SZ represents the signal entropy value; the medium interleaving coefficient Cinter is calculated as follows: Where SZ0 represents the signal entropy value under uniform material.

[0018] In a preferred embodiment, the recovery time information of the ultrasonic probe is obtained, the recovery time fluctuation degree of the ultrasonic probe is analyzed, and the probe recovery time fluctuation coefficient is calculated to measure the recovery time fluctuation degree of the ultrasonic probe;

[0019] The logic for obtaining the probe recovery time fluctuation coefficient is as follows:

[0020] By recording the ultrasonic pulse emission time of each ultrasonic probe, the time interval between adjacent ultrasonic pulse emissions is used as the probe recovery time to obtain a set of probe recovery time data sets TJ = {tj j}={tj 1 ,tj 2 ,...,tj m}, where tj j represents the probe recovery time of the jth acquisition, j = {1, 2, ..., m}, m is a positive integer; calculate the mean value of the probe recovery time μtj, the expression is as follows: Calculate the standard deviation of the probe recovery time σtj, the expression is as follows: Calculate the long-term dependence factor Ht of the probe recovery time data set, the expression is as follows: Where JC represents the range of the probe recovery time data set; the probe recovery time fluctuation coefficient Cvdsa is calculated as follows:

[0021] In a preferred embodiment, the environmental interference information of the ultrasonic signal is obtained, the degree of environmental interference of the ultrasonic signal is analyzed, and the environmental interference coefficient is calculated to measure the degree of influence of the environmental interference on the ultrasonic signal;

[0022] The logic for obtaining the environmental interference coefficient is as follows:

[0023] The echo signal x(t) from the thin-wall area is obtained, and the echo signal is converted into the time-frequency domain using Fourier transform to obtain the time-frequency spectrum of the echo signal, as follows: Where X(t,f) is the time-frequency spectrum of the echo signal, which represents the frequency distribution at time t and frequency f, x(τ) represents the original echo signal, ω(t-τ) represents the window function, and e -j2πfτ represents the Fourier basis function, t is the time representing the center point of the analysis window, and f represents the frequency. The interference noise energy is calculated according to the time-frequency spectrum of the echo signal. The expression is as follows: Where Enoise represents the interference noise energy, f noise,low represents the low frequency threshold of noise, f noise,high Represents the high-frequency threshold of noise; inputs the time-frequency spectrum of the echo signal into the convolutional neural network to output the probability of the environmental interference category; calculates the environmental interference coefficient C env , the expression is as follows: Where Pclass g Indicates the probability of the g-th environmental interference category, Enoise g Represents the interference noise energy of the g-th environmental interference category.

[0024] In a preferred embodiment, a casing forging missed detection risk assessment model is constructed based on the medium interleaving coefficient, the probe recovery time fluctuation coefficient, and the environmental interference coefficient, and the casing forging missed detection risk assessment index Rleak is output. The model is based on the following formula: Where, Cinter represents the medium interleaving coefficient, Cvdsa represents the probe recovery time fluctuation coefficient, and C env represents the environmental interference coefficient, h 1 、h 2 、h 3 They represent the preset proportional coefficients of the medium interleaving coefficient, the probe recovery time fluctuation coefficient, and the environmental interference coefficient, respectively, and h 1 、h 2 、h 3 Both are greater than 0.

[0025] In a preferred embodiment, the casing forging missed inspection risk assessment index is compared with a preset casing forging missed inspection risk assessment index threshold, and the missed inspection risk of the aircraft engine casing forging is divided as follows:

[0026] If the casing forging missed inspection risk assessment index is greater than the casing forging missed inspection risk assessment index threshold, a missed inspection risk signal is generated; if the casing forging missed inspection risk assessment index is less than or equal to the casing forging missed inspection risk assessment index threshold, there is no need to generate a missed inspection risk signal.

[0027] Technical effects and advantages of the present invention:

[0028] The present invention calculates the thickness of a casing forging according to the ultrasonic propagation time, identifies the thin-walled area according to the thickness of the casing forging, obtains the medium interlacing information of the thin-walled area, and calculates the medium interlacing coefficient according to the medium interlacing information of the thin-walled area to evaluate the medium interlacing degree of the thin-walled area; obtains the recovery time information of the ultrasonic probe, and calculates the probe recovery time fluctuation coefficient according to the recovery time information of the ultrasonic probe to evaluate the recovery time fluctuation degree of the ultrasonic probe; obtains the environmental interference information of the ultrasonic signal, and calculates the environmental interference coefficient according to the environmental interference information of the ultrasonic signal to evaluate the influence of the environmental interference on the ultrasonic signal; through quantitative analysis of the medium interlacing degree of the thin-walled area, the probe recovery time fluctuation degree and the environmental interference, it can effectively identify the potential risk of missed detection, avoid misjudgment and missed detection caused by the difficulty of traditional detection methods in handling complex surface morphology, and through precise ultrasonic detection and multi-dimensional risk assessment models, significantly improve the detection efficiency and accuracy of aircraft engine casing forgings, reduce the risk of missed detection, ensure the structural safety and reliability of the aircraft engine, and ensure the high performance and safe operation of the aircraft. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to facilitate understanding by those skilled in the art, the present invention is further described below in conjunction with the accompanying drawings;

[0030] Figure 1 The figure is a flow chart of a method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0031] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0032] Example: Figure 1 The invention provides an automated production and intelligent detection method for an aircraft engine casing forging, comprising the following steps:

[0033] Step S1, transmitting ultrasonic pulses to different positions of the casing forging, and using a receiver to capture echo signals to measure ultrasonic propagation time, calculating the thickness of the casing forging according to the ultrasonic propagation time, and identifying the thin-walled area according to the thickness of the casing forging;

[0034] Step S2, obtaining medium interleaving information of the thin-walled area and calculating a medium interleaving coefficient according to the medium interleaving information of the thin-walled area to evaluate the medium interleaving degree of the thin-walled area;

[0035] Step S3, obtaining recovery time information of the ultrasonic probe, and calculating the probe recovery time fluctuation coefficient according to the recovery time information of the ultrasonic probe, so as to evaluate the recovery time fluctuation degree of the ultrasonic probe;

[0036] Step S4, obtaining environmental interference information of the ultrasonic signal and calculating an environmental interference coefficient according to the environmental interference information of the ultrasonic signal to evaluate the influence of the environmental interference on the ultrasonic signal;

[0037] Step S5, constructing a casing forging missed inspection risk assessment model according to the medium interleaving coefficient, the probe recovery time fluctuation coefficient, and the environmental interference coefficient, outputting a casing forging missed inspection risk assessment index, and assessing the missed inspection risk of the aircraft engine casing forging;

[0038] Step S1, by transmitting ultrasonic pulses to different positions of the casing forging, recording the ultrasonic pulse transmission time, using a receiver to capture the echo signal and record the echo signal capture time, calculating the ultrasonic propagation time according to the ultrasonic pulse transmission time and the echo signal capture time, the expression is as follows: csb = |hbt-fst|, where csb represents the ultrasonic propagation time, hbt represents the echo signal capture time, and fst represents the ultrasonic pulse transmission time; the casing forging thickness is calculated according to the ultrasonic propagation time, the expression is as follows: Where jxd represents the thickness of the casing forging, vs represents the sound velocity of the casing forging material;

[0039] Compare the thickness of the casing forging with a preset standard thickness, and mark the area where the thickness of the casing forging is less than the standard thickness as a thin-walled area;

[0040] It should be noted that the sound velocity of the casing forging material refers to the speed at which the ultrasonic wave propagates in the material. The sound velocity is mainly affected by the elastic modulus, density and other physical properties of the material. The propagation time of the ultrasonic wave can be measured by a sample experiment with a known thickness, and the sound velocity of the material can be inferred. For commonly used materials (such as steel, aluminum, titanium alloy, etc.), the sound velocity values ​​given in the standard data can be referred to.

[0041] It should be noted that the standard thickness can be preset by technicians in this field according to the design requirements and actual process of the aircraft engine casing forgings, and will not be described in detail here;

[0042] Step S2, obtaining medium interleaving information of the thin-walled area and calculating a medium interleaving coefficient according to the medium interleaving information of the thin-walled area to evaluate the medium interleaving degree of the thin-walled area;

[0043] In this embodiment, the medium interlacing coefficient is used to measure the complexity of the distribution of heterogeneous interfaces (such as grain boundaries, inclusions, phase interfaces, etc.) inside the aeroengine casing forgings. These heterogeneous interfaces will cause scattering, reflection and mode conversion during ultrasonic propagation, affecting the propagation characteristics of the signal, especially the detection of material defects. The more heterogeneous interfaces there are inside the material, the stronger the scattering degree of the ultrasonic signal, resulting in an increase in the diversity of the echo signal. By quantifying the energy distribution entropy of the ultrasonic backscattered signal, the degree of medium interlacing can be quantitatively evaluated, so as to understand the structural complexity of the material and the difficulty of detection; the risk of missed detection of aeroengine casing forgings is evaluated based on the medium interlacing coefficient, which can significantly improve the accuracy and reliability of detection. The medium interlacing coefficient is to quantify the influence of factors such as grain boundaries, inclusions and heterogeneous interfaces inside the material on ultrasonic propagation by analyzing the energy distribution entropy of the ultrasonic echo signal. When the medium interlacing degree of the material is high, the propagation of the ultrasonic signal will be subject to stronger scattering and attenuation, resulting in an increase in the randomness of the signal, and it is difficult to accurately identify small defects. By quantifying the medium interleaving coefficient, it is possible to identify in advance those areas that may be difficult to detect, especially for thin-walled areas or casing forgings with complex structures. It is possible to accurately assess the difficulty of detection and adjust the frequency, power and other parameters of the ultrasonic probe according to the test results to optimize the detection process. This not only helps to improve the defect detection capabilities in thin-walled areas and areas with complex geometric shapes, but also effectively reduces the occurrence of missed detections, especially when traditional detection methods may have limitations. It can provide real-time risk warnings during the production process, reduce missed detection problems caused by material complexity, ensure the structural integrity and safety of aircraft engine casing forgings, and thus provide more efficient guarantees for quality control of the entire production process.

[0044] By obtaining the medium interleaving information of the thin-walled area, analyzing the medium interleaving degree of the thin-walled area, and calculating the medium interleaving coefficient, the medium interleaving degree of the thin-walled area is measured;

[0045] The logic for obtaining the medium interleaving coefficient is as follows:

[0046] Get the echo signal x(t) from the thin-wall area and decompose the echo signal using wavelet transform, as follows: Where W ψ (a,b) represents the wavelet transform coefficient, a represents the scale factor, b represents the translation factor, ψ represents the mother wavelet function, and x(t) represents the echo signal at time t. The echo signal from the thin-walled area is divided into N time domain windows of the same interval, and the signal energy of each time domain window is calculated. The expression is as follows: E i =∑ a,b |W ψ (a,b)| 2 , where E iRepresents the signal energy of the i-th time domain window; calculate the signal energy ratio, the expression is as follows: Where P i Represents the signal energy proportion of the i-th time domain window, i={1,2,...,N}, N is a positive integer; calculate the signal entropy value, the expression is as follows: Where SZ represents the signal entropy value; the medium interleaving coefficient Cinter is calculated as follows: Where SZ0 represents the signal entropy value under uniform material;

[0047] Step S3, obtaining recovery time information of the ultrasonic probe, and calculating the probe recovery time fluctuation coefficient according to the recovery time information of the ultrasonic probe, so as to evaluate the recovery time fluctuation degree of the ultrasonic probe;

[0048] The recovery time fluctuation of the ultrasonic probe is caused by factors such as the physical properties of the probe, damping effect, electronic circuit delay, and time difference between signal transmission and reception. These factors cause the probe to need a certain recovery time before it can transmit a new signal again after the ultrasonic signal is transmitted. If the fluctuation of the probe recovery time is too large, it will aggravate the expansion of the near-surface blind area, thereby affecting the detection accuracy and integrity of the thin-walled area; in this implementation, the probe recovery time fluctuation coefficient is used to measure the degree of fluctuation of the ultrasonic probe recovery time. Based on the probe recovery time fluctuation coefficient, the risk of missed detection of aircraft engine casing forgings is evaluated, which can effectively identify the detection errors caused by unstable probe performance or external factors. The recovery time fluctuation coefficient reflects the response stability of the ultrasonic probe, which is crucial for accurately detecting defects in thin-walled areas. The recovery time fluctuation is caused by factors such as the physical properties of the probe (such as damping characteristics, frequency response) and circuit delay. Excessive fluctuations will cause the probe to fail to recover to the next measurement cycle in a timely or accurate manner after receiving the echo signal, thereby affecting the capture of the ultrasonic signal. By accurately quantifying the degree of fluctuation of the recovery time, we can evaluate whether the probe is at risk of performance degradation and compensate for its impact by adjusting the ultrasonic signal acquisition strategy. Specifically, when the recovery time fluctuation coefficient is high, it means that the recovery performance of the probe is poor and the signal recovery delay is large, which may lead to the expansion of the near-surface blind area. The detected signal may not reflect the real defect, thereby increasing the probability of missed detection or misjudgment. This fluctuation often occurs in the high-frequency response area, especially in thin-walled and complex-shaped casing forgings, which affects the accuracy of detection. By incorporating the recovery time fluctuation coefficient into the missed detection risk assessment, the risk of poor probe performance can be identified in advance, and targeted adjustment measures can be taken, such as increasing the calibration frequency, optimizing the signal recovery time, selecting a suitable ultrasonic probe, and increasing the monitoring of the probe position, thereby effectively reducing the missed detection rate and ensuring the safety and structural integrity of aircraft engine casing forgings. This evaluation method combines the real-time changes in probe performance and the actual production environment, making the detection process more intelligent and reliable, thereby greatly improving the overall detection quality and reducing hidden dangers in the production process.

[0049] By acquiring the recovery time information of the ultrasonic probe, analyzing the fluctuation degree of the recovery time of the ultrasonic probe, and calculating the probe recovery time fluctuation coefficient, the fluctuation degree of the recovery time of the ultrasonic probe is measured;

[0050] The logic for obtaining the probe recovery time fluctuation coefficient is as follows:

[0051] By recording the ultrasonic pulse emission time of each ultrasonic probe, the time interval between adjacent ultrasonic pulse emissions is used as the probe recovery time to obtain a set of probe recovery time data sets TJ = {tj j}={tj 1 ,tj2 ,...,tj m}, where tj j represents the probe recovery time of the jth acquisition, j = {1, 2, ..., m}, m is a positive integer; calculate the mean value of the probe recovery time μtj, the expression is as follows: Calculate the standard deviation of the probe recovery time σtj, the expression is as follows: Calculate the long-term dependence factor Ht of the probe recovery time data set, the expression is as follows: Where JC represents the range of the probe recovery time data set, that is, the difference between the maximum and minimum values ​​of the probe recovery time in the probe recovery time data set; the probe recovery time fluctuation coefficient Cvdsa is calculated, and the expression is as follows:

[0052]

[0053] Step S4, obtaining environmental interference information of the ultrasonic signal and calculating an environmental interference coefficient according to the environmental interference information of the ultrasonic signal to evaluate the influence of the environmental interference on the ultrasonic signal;

[0054] In this embodiment, the environmental interference coefficient is used to measure the degree of influence of environmental noise interference on ultrasonic signals. In the ultrasonic detection process of aircraft engine casing forgings, external environmental factors such as electromagnetic interference (such as frequency converter harmonics) and mechanical vibration (such as workshop background noise) may be coupled into the ultrasonic signal, thereby affecting the quality of the signal. Electromagnetic interference and mechanical noise can cause the ultrasonic signal to be distorted, attenuated or distorted, resulting in a decrease in the accuracy of the detection results. By evaluating the intensity of environmental interference and quantifying the ratio of signal to noise, the impact of the environment on ultrasonic detection can be effectively evaluated. The risk of missed detection of aircraft engine casing forgings is evaluated based on the environmental interference coefficient, which can effectively identify and quantify the negative impact from the external environment (such as electromagnetic interference, mechanical vibration, etc.). These environmental factors may cause serious interference to the ultrasonic detection system, thereby affecting the accuracy and reliability of the detection results. In the ultrasonic detection process of aircraft engine casing forgings, environmental interference is often an important factor leading to missed detection and misjudgment. Electromagnetic interference, such as high-frequency noise generated by frequency converters, welding equipment, etc., may cause the ultrasonic probe to receive unnecessary signals, and mechanical vibration (such as workshop background noise, vibration of production equipment) will also increase signal noise, affecting the accurate capture of ultrasonic signals. Especially in the thin-walled area of ​​the casing forging, the signal itself is weak. In the case of large external interference, the defect signal may be misjudged as noise or completely undetectable, resulting in missed detection. Missed detection risk assessment based on environmental interference coefficient can not only improve the accuracy of detection, but also provide a more scientific basis for quality control of the production process, help reduce the flow of unqualified products into the market, and ensure the safety and reliability of aircraft engine casing forgings. The introduction of this evaluation method enables the ultrasonic detection system to adapt to complex production environments in real time, providing guarantees for high-precision and high-reliability aviation manufacturing.

[0055] By acquiring the environmental interference information of the ultrasonic signal, analyzing the degree of environmental interference of the ultrasonic signal, and calculating the environmental interference coefficient, the influence of environmental interference on the ultrasonic signal is measured;

[0056] The logic for obtaining the environmental interference coefficient is as follows:

[0057] The echo signal x(t) from the thin-wall area is obtained, and the echo signal is converted into the time-frequency domain using Fourier transform to obtain the time-frequency spectrum of the echo signal, as follows: Where X(t,f) is the time-frequency spectrum of the echo signal, which represents the frequency distribution at time t and frequency f, x(τ) represents the original echo signal, ω(t-τ) represents the window function, which is used for local processing of the echo signal, and e -j2πfτ Represents the Fourier basis function, which is used to extract the frequency component of the echo signal. t is the time representing the center point of the analysis window, and f is the frequency. The interference noise energy is calculated based on the time-frequency spectrum of the echo signal. The expression is as follows: Where Enoise represents the interference noise energy, f noise,low represents the low frequency threshold of noise, f noise,high Represents the high-frequency threshold of noise; inputs the time-frequency spectrum of the echo signal into the convolutional neural network to output the probability of the environmental interference category; calculates the environmental interference coefficient C env , the expression is as follows: Where Pclass g Indicates the probability of the g-th environmental interference category, Enoise g represents the interference noise energy of the g-th environmental interference category;

[0058] It should be noted that convolutional neural network is a powerful tool for processing time-frequency graph data. It can be used to automatically extract features and classify them, and identify different areas of noise and echo signals. The convolutional neural network consists of a convolutional layer, a pooling layer, and a fully connected layer. The input of the convolutional neural network is the time-frequency graph of the echo signal, and the output is the probability of the environmental interference category. The model training is performed using an annotated echo signal data set to identify the environmental interference noise characteristics in the echo signal, and the probability of the environmental interference category is output based on the trained convolutional neural network.

[0059] Step S5, constructing a casing forging missed inspection risk assessment model according to the medium interleaving coefficient, the probe recovery time fluctuation coefficient, and the environmental interference coefficient, outputting a casing forging missed inspection risk assessment index, and assessing the missed inspection risk of the aircraft engine casing forging;

[0060] According to the medium interleaving coefficient, the probe recovery time fluctuation coefficient, and the environmental interference coefficient, a risk assessment model for missed detection of casing forgings is constructed, and the risk assessment index for missed detection of casing forgings Rleak is output. The formula based on the model is as follows Where, Cinter represents the medium interleaving coefficient, Cvdsa represents the probe recovery time fluctuation coefficient, and C env represents the environmental interference coefficient, h 1 、h 2 、h 3 They represent the preset proportional coefficients of the medium interleaving coefficient, the probe recovery time fluctuation coefficient, and the environmental interference coefficient, respectively, and h 1 、h 2 、h 3 All are greater than 0;

[0061] It should be noted that before constructing the risk assessment model for missed detection of casing forgings, it is necessary to ensure that the medium interleaving coefficient, the probe recovery time fluctuation coefficient, and the environmental interference coefficient are all normalized. Commonly used normalization methods include Min-Max normalization and Z-Score normalization. 1 、h 2 、h 3Set it according to the actual situation. For example, adopt the expert empowerment method, that is, invite experts in relevant fields to determine the preset proportion coefficients of various indicators through professional opinion surveys and comprehensive evaluations;

[0062] It can be seen from the above calculation expression that the larger the medium interleaving coefficient, the larger the probe recovery time fluctuation coefficient, and the larger the environmental interference coefficient, the larger the risk assessment index of missed detection of casing forgings, which means that the effect of ultrasonic detection is poor and there is a high risk of missed detection. On the contrary, the smaller the medium interleaving coefficient, the smaller the probe recovery time fluctuation coefficient, and the smaller the environmental interference coefficient, the smaller the risk assessment index of missed detection of casing forgings, which means that the detection process is more stable and the probability of missed detection is lower.

[0063] The risk assessment model for missed detection of casing forgings is constructed based on the medium interleaving coefficient, the probe recovery time fluctuation coefficient, and the environmental interference coefficient. By quantifying and comprehensively considering these factors, the model can provide a comprehensive and systematic assessment tool to help accurately predict the risk of missed detection of aircraft engine casing forgings during ultrasonic testing. The medium interleaving coefficient reflects the complexity of the internal structure of the material and its impact on the detection signal; the probe recovery time fluctuation coefficient reveals the impact of the stability and reaction speed of the ultrasonic probe on the detection accuracy; and the environmental interference coefficient reveals the interference of external noise, vibration, etc. on the detection results. Through the comprehensive evaluation of these three key factors, the model can accurately measure the potential risk of missed detection, thereby providing a basis for the optimization of the detection process in the production process, helping producers to promptly identify and respond to potential detection errors, especially in the detection of complex structures such as thin-walled areas, effectively reducing the risk of missed detection and improving overall production quality and safety.

[0064] The missed inspection risk assessment index of casing forgings is compared with the preset missed inspection risk assessment index threshold of casing forgings, and the missed inspection risk of aircraft engine casing forgings is divided as follows:

[0065] If the casing forging missed detection risk assessment index is greater than the casing forging missed detection risk assessment index threshold, a missed detection risk signal is generated, indicating that the aircraft engine casing forging has a high risk of missed detection during the ultrasonic detection process, and potential defects or flaws may not be accurately detected. Therefore, further measures need to be taken to manage risks, such as: Re-evaluate the detection settings: optimize the configuration, detection parameters, signal processing, etc. of the ultrasonic probe, adjust the detection method or select different types of probes to improve the sensitivity and accuracy of the detection. Strengthen environmental control: Strengthen the control of the production or testing environment, reduce the impact of electromagnetic interference, mechanical vibration and other factors on the ultrasonic signal, and ensure signal quality. Further manual inspection or subsequent inspection: For high-risk casing forgings, manual auxiliary inspection or other non-destructive detection methods (such as X-ray detection, CT scanning, etc.) can be performed for supplementary inspection to ensure that defects can be found. Quality review and process optimization: Start the quality review process, conduct in-depth analysis and improvement of technical defects or management loopholes that may exist in the production and testing process, and ensure product quality and safety.

[0066] If the casing forging missed detection risk assessment index is less than or equal to the casing forging missed detection risk assessment index threshold, there is no need to generate a missed detection risk signal, indicating that the ultrasonic detection process of the aircraft engine casing forging is within the normal range, the missed detection risk is low, and the detection result is reliable. At this time, it means that the accuracy and reliability of the current detection process have reached the expected standard, and no potential large missed detection risk has been found.

[0067] Specifically, this means:

[0068] The test results are reliable: The impact of various factors in the ultrasonic testing process (such as medium interleaving, probe stability and environmental interference) on the test results is within a controllable range, so the test results can be trusted, and the quality of aircraft engine casing forgings meets the standards. No further intervention is required: At this risk level, the production line and testing process do not require additional adjustments or interventions to ensure production efficiency and normal operation, and continue to follow the established process. Continuous monitoring and optimization: Although the current test has not found any risk of missed detection, the testing process should still be monitored and optimized regularly to ensure that the test quality remains at a high level as the production environment changes or technology advances. Maintain existing testing standards: The current testing equipment, parameters and environmental settings can continue to be used, and there is no need to immediately change the testing process or add additional safety checks.

[0069] The present invention calculates the thickness of a casing forging according to the ultrasonic propagation time, identifies the thin-walled area according to the thickness of the casing forging, obtains the medium interlacing information of the thin-walled area, and calculates the medium interlacing coefficient according to the medium interlacing information of the thin-walled area to evaluate the medium interlacing degree of the thin-walled area; obtains the recovery time information of the ultrasonic probe, and calculates the probe recovery time fluctuation coefficient according to the recovery time information of the ultrasonic probe to evaluate the recovery time fluctuation degree of the ultrasonic probe; obtains the environmental interference information of the ultrasonic signal, and calculates the environmental interference coefficient according to the environmental interference information of the ultrasonic signal to evaluate the influence of the environmental interference on the ultrasonic signal; through quantitative analysis of the medium interlacing degree of the thin-walled area, the probe recovery time fluctuation degree and the environmental interference, it can effectively identify the potential risk of missed detection, avoid misjudgment and missed detection caused by the difficulty of traditional detection methods in handling complex surface morphology, and through precise ultrasonic detection and multi-dimensional risk assessment models, significantly improve the detection efficiency and accuracy of aircraft engine casing forgings, reduce the risk of missed detection, ensure the structural safety and reliability of the aircraft engine, and ensure the high performance and safe operation of the aircraft.

[0070] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0071] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0072] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. The automated production and intelligent detection method of aircraft engine casing forgings is characterized by: The steps include: Step S1, transmitting ultrasonic pulses to different positions of the casing forging, and using a receiver to capture echo signals to measure ultrasonic propagation time, calculating the thickness of the casing forging according to the ultrasonic propagation time, and identifying the thin-walled area according to the thickness of the casing forging; Step S2, obtaining medium interleaving information of the thin-walled area and calculating a medium interleaving coefficient according to the medium interleaving information of the thin-walled area to evaluate the medium interleaving degree of the thin-walled area; Step S3, evaluating the recovery time fluctuation degree of the ultrasonic probe by acquiring the recovery time information of the ultrasonic probe and calculating the probe recovery time fluctuation coefficient according to the recovery time information of the ultrasonic probe; Step S4, obtaining environmental interference information of the ultrasonic signal and calculating an environmental interference coefficient according to the environmental interference information of the ultrasonic signal to evaluate the influence of the environmental interference on the ultrasonic signal; Step S5, constructing a casing forging missed inspection risk assessment model according to the medium interleaving coefficient, the probe recovery time fluctuation coefficient, and the environmental interference coefficient, outputting a casing forging missed inspection risk assessment index, and assessing the missed inspection risk of aircraft engine casing forgings.

2. The automated production and intelligent detection method for aircraft engine casing forgings according to claim 1, characterized in that: In step S1, ultrasonic pulses are emitted to different positions of the casing forging, the ultrasonic pulse emission time is recorded, the echo signal is captured by a receiver and the echo signal capture time is recorded, and the ultrasonic propagation time is calculated according to the ultrasonic pulse emission time and the echo signal capture time. The expression is as follows: csb = |hbt-fst|, where csb represents the ultrasonic propagation time, hbt represents the echo signal capture time, and fst represents the ultrasonic pulse emission time; the casing forging thickness is calculated according to the ultrasonic propagation time. The expression is as follows: Where jxd represents the thickness of the casing forging, vs represents the sound velocity of the casing forging material; The thickness of the casing forging is compared with a preset standard thickness, and the area where the thickness of the casing forging is less than the standard thickness is marked as a thin-walled area.

3. The automated production and intelligent detection method for aircraft engine casing forgings according to claim 1, characterized in that: By obtaining the medium interleaving information of the thin-walled area, analyzing the medium interleaving degree of the thin-walled area, and calculating the medium interleaving coefficient, the medium interleaving degree of the thin-walled area is measured; The logic for obtaining the medium interleaving coefficient is as follows: Get the echo signal x(t) from the thin-wall area and decompose the echo signal using wavelet transform, as follows: Where W ψ (a,b) represents the wavelet transform coefficient, a represents the scale factor, b represents the translation factor, ψ represents the mother wavelet function, and x(t) represents the echo signal at time t. The echo signal from the thin-walled area is divided into N time domain windows of the same interval, and the signal energy of each time domain window is calculated. The expression is as follows: E i =∑ a,b |W ψ (a,b)| 2 , where E i Represents the signal energy of the i-th time domain window; calculate the signal energy ratio, the expression is as follows: Where P i represents the signal energy proportion of the i-th time domain window, i = {1, 2, ..., N}, N is a positive integer; Calculate the signal entropy value, the expression is as follows: Where SZ represents the signal entropy value; the medium interleaving coefficient Cinter is calculated as follows: Where SZ0 represents the signal entropy value under uniform material.

4. The automated production and intelligent detection method for aircraft engine casing forgings according to claim 1, characterized in that: By acquiring the recovery time information of the ultrasonic probe, analyzing the fluctuation degree of the recovery time of the ultrasonic probe, and calculating the probe recovery time fluctuation coefficient, the fluctuation degree of the recovery time of the ultrasonic probe is measured; The logic for obtaining the probe recovery time fluctuation coefficient is as follows: By recording the ultrasonic pulse emission time of each ultrasonic probe, the time interval between adjacent ultrasonic pulse emissions is used as the probe recovery time to obtain a set of probe recovery time data sets TJ = {tj j }={tj1,tj2,...,tj m }, where tj j represents the probe recovery time of the jth acquisition, j = {1, 2, ..., m}, m is a positive integer; calculate the mean value of the probe recovery time μtj, the expression is as follows: Calculate the standard deviation of the probe recovery time σtj, the expression is as follows: Calculate the long-term dependence factor Ht of the probe recovery time data set, the expression is as follows: Where JC represents the range of the probe recovery time data set; the probe recovery time fluctuation coefficient Cvdsa is calculated as follows:

5. The automated production and intelligent detection method for aircraft engine casing forgings according to claim 1, characterized in that: By acquiring the environmental interference information of the ultrasonic signal, analyzing the degree of environmental interference of the ultrasonic signal, and calculating the environmental interference coefficient, the influence of environmental interference on the ultrasonic signal is measured; The logic for obtaining the environmental interference coefficient is as follows: The echo signal x(t) from the thin-wall area is obtained, and the echo signal is converted into the time-frequency domain using Fourier transform to obtain the time-frequency spectrum of the echo signal, as follows: Where X(t,f) is the time-frequency spectrum of the echo signal, which represents the frequency distribution at time t and frequency f, x(τ) represents the original echo signal, ω(t-τ) represents the window function, and e -j2πfτ represents the Fourier basis function, t is the time representing the center point of the analysis window, and f represents the frequency. The interference noise energy is calculated according to the time-frequency spectrum of the echo signal. The expression is as follows: Where Enoise represents the interference noise energy, f noise,low represents the low frequency threshold of noise, f noise,high Represents the high-frequency threshold of noise; inputs the time-frequency spectrum of the echo signal into the convolutional neural network to output the probability of the environmental interference category; calculates the environmental interference coefficient C env , the expression is as follows: Where Pclass g Indicates the probability of the g-th environmental interference category, Enoise g Represents the interference noise energy of the g-th environmental interference category.

6. The automated production and intelligent detection method for aircraft engine casing forgings according to claim 1, characterized in that: According to the medium interleaving coefficient, the probe recovery time fluctuation coefficient, and the environmental interference coefficient, a risk assessment model for missed detection of casing forgings is constructed, and the risk assessment index for missed detection of casing forgings Rleak is output. The formula based on the model is as follows Where, Cinter represents the medium interleaving coefficient, Cvdsa represents the probe recovery time fluctuation coefficient, and C env represents the environmental interference coefficient, h1, h2, and h3 represent the preset proportional coefficients of the medium interleaving coefficient, the probe recovery time fluctuation coefficient, and the environmental interference coefficient, respectively, and h1, h2, and h3 are all greater than 0.

7. The automated production and intelligent detection method for aircraft engine casing forgings according to claim 6, characterized in that: The missed inspection risk assessment index of casing forgings is compared with the preset missed inspection risk assessment index threshold of casing forgings, and the missed inspection risk of aircraft engine casing forgings is divided as follows: If the casing forging missed inspection risk assessment index is greater than the casing forging missed inspection risk assessment index threshold, a missed inspection risk signal is generated; if the casing forging missed inspection risk assessment index is less than or equal to the casing forging missed inspection risk assessment index threshold, there is no need to generate a missed inspection risk signal.

Images