Ultrasonic nondestructive testing A scanning signal noise reduction method and system based on improved U-NET

Abstract

The invention discloses an ultrasonic non-destructive testing A scanning signal noise reduction method and system based on an improved U-NET, and belongs to the technical field of industrial ultrasonic non-destructive testing. A de-noising neural network based on U-Net is established; and training a denoising neural network based on U-Net by using the preprocessed ultrasonic A scanning defect signal, and then inputting an original signal with noise into the trained neural network to realize noise reduction processing. According to the method, the stability, robustness and universality of the noise reduction model can be improved, and various kinds of noise can be effectively denoised.

Description

Background Art

[0002] As an emerging material, carbon fiber composites have the characteristics of high tensile strength, high elastic modulus, low density, and strong conductivity, and are widely used in the aerospace field. The complex manufacturing process of carbon fiber composites also determines that various defects will be generated inside the material due to temperature gradients and residual stresses during the production process. In order to avoid serious consequences caused by the expansion of internal defects, the evaluation and testing of carbon fiber composites during the production process and under service conditions has become the primary task before the current work.

[0003] As a non-contact measurement method, non-destructive testing has been widely used in the processing and manufacturing industry. Ultrasonic non-destructive testing has become the mainstream testing method due to its high sensitivity, good directivity and wide applicability. With the rapid development of water immersion ultrasonic testing technology, the automation level, detection speed and accuracy of ultrasonic non-destructive testing technology have been greatly improved.

[0004] Ultrasonic A-scan is a commonly used nondestructive testing technology. In the A-scan signal, the horizontal axis represents the propagation time or propagation distance of the ultrasonic wave in the object being tested, and the vertical axis represents the amplitude of the ultrasonic reflection wave, that is, the signal strength reflected by the ultrasonic wave when it encounters different media inside the object being tested. Therefore, the A-scan signal is the most original signal in ultrasonic testing, and the subsequent defect detection steps are closely related to the A-scan signal.

[0005] In the process of defect detection of carbon fiber composite materials, due to its anisotropy, harsh detection environment and the instrument's own factors, the A-scan signal inevitably has large noise interference and signal attenuation. Therefore, for non-destructive testing by directly judging the distribution of curves, the inspection personnel need to have a strong professional knowledge base and rich inspection experience, and the object under test needs to be inspected multiple times; while directly using the noisy A-scan signal for two-dimensional image reconstruction will lead to negative effects such as image distortion and poor reconstruction accuracy.

[0006] The existing mainstream methods for ultrasonic A-scan signal noise reduction mainly include wavelet packet decomposition and CEEMD decomposition. However, the wavelet packet decomposition method has high computational complexity, large amount of calculation, overfitting risk, and poor effect on certain types of noise, and cannot remove different types of noise. The CEEMD decomposition method has poor stability, relatively more complex calculation process and higher professional requirements for researchers. When removing noise, useful signals may be removed at the same time. There is also the problem that the noise reduction effect is only good for specific noise. Summary of the invention

[0007] The technical problem to be solved by the present invention is to provide a method and system for denoising an A-scan signal of ultrasonic nondestructive testing based on an improved U-NET in response to the deficiencies in the above-mentioned prior art, so as to achieve denoising of the ultrasonic A-scan signal in a noisy background and separate a clean signal, so as to solve the technical problem that the current method only targets specific noise and cannot remove various types of different noises.

[0008] The present invention adopts the following technical solutions:

[0009] A method for reducing noise of ultrasonic nondestructive testing A-scan signals based on improved U-NET, comprising the following steps:

[0010] S1, preprocessing the ultrasonic A-scan signal;

[0011] S2. Build a denoising neural network based on U-Net;

[0012] S3, using the ultrasonic A-scan signal obtained in step S1 to train the U-Net-based denoising neural network obtained in step S2, and then inputting the original signal with noise into the trained neural network to achieve noise reduction processing.

[0013] Preferably, step S1 specifically includes:

[0014] Gaussian noise of different decibels is added to the noise-free ultrasound A-scan signal and then normalized.

[0015] Preferably, in step S2, the U-Net-based denoising neural network includes an encoding part, a central part, a decoding part and an output layer, and the features are converted into the final number of output channels without changing the feature dimension, and a skip connection layer is used to fuse the feature maps in the downsampling process in the upsampling process;

[0016] The encoding part includes two convolutional layers, two batch normalization layers, and two maximum pooling layers;

[0017] The central part includes two convolutional layers and two maximum pooling layers;

[0018] The decoding part includes two transposed convolutional layers, two convolutional layers, and two batch normalization layers;

[0019] The output layer consists of a point-wise convolutional layer.

[0020] Preferably, the activation function between the convolution layer and the normalization layer is the PRelu activation function as follows:

[0021]

[0022] Among them, α is a learnable parameter and x is the input of the activation function.

[0023] Preferably, the convolution kernel space size in the coding layer is 3*1 elements, and the step size is 1 element;

[0024] The spatial size of the maximum pooling layer kernel is 2*1 elements, and the stride is 2 elements;

[0025] The convolution kernel space size in the center layer is 3*1 element, and the stride is 1 element;

[0026] The spatial size of the transposed convolution kernel in the decoder layer is 2*1 elements, and the stride is 2 elements;

[0027] The spatial size of the convolution kernel is 3*1 element, and the stride is 1 element.

[0028] Preferably, the U-Net-based denoising neural network also includes a channel attention mechanism and a spatial attention mechanism; the channel attention mechanism and the spatial attention mechanism are connected in series to form a CABM module, which is then inserted into the U-Net-based denoising neural network.

[0029] Preferably, the channel attention mechanism performs global average pooling and global maximum pooling on the feature map to reduce the dimension, obtains the results of average pooling and maximum pooling respectively, and then inputs them into a shared multi-layer perceptron for learning. The output of the multi-layer perceptron is compressed by a sigmoid activation function to obtain the weight of each channel, and the contribution of the specified channel is enhanced or weakened by the weight.

[0030] The spatial attention mechanism dynamically adjusts the importance of different spatial positions of the input data in the U-Net based denoising neural network.

[0031] Preferably, in step S3, the U-Net-based denoising neural network is trained using a mean square error loss function and an Adam optimizer.

[0032] Preferably, the mean square error loss function MSE is:

[0033]

[0034] Where N is the number of samples, x i is the predicted value, y i is the true value.

[0035] In a second aspect, an embodiment of the present invention provides an ultrasonic nondestructive testing A-scan signal noise reduction system based on an improved U-NET, comprising:

[0036] A preprocessing module, which preprocesses the ultrasonic A-scan signal;

[0037] Network module, build a denoising neural network based on U-Net;

[0038] The denoising module uses the preprocessed ultrasonic A-scan signal to train the U-Net-based denoising neural network, and then inputs the noisy original signal into the trained neural network to achieve denoising processing.

[0039] In a third aspect, a computer device comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the steps of the above-mentioned method for denoising an A-scan signal of ultrasonic nondestructive testing based on an improved U-NET are implemented.

[0040] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, including a computer program, which, when executed by a processor, implements the steps of the above-mentioned ultrasonic non-destructive testing A-scan signal denoising method based on improved U-NET.

[0041] Compared with the prior art, the present invention has at least the following beneficial effects:

[0042] A method for reducing the noise of A-scan signals in ultrasonic nondestructive testing based on improved U-NET. When using ultrasound to perform nondestructive testing on composite materials, due to the anisotropic nature of the composite materials, there is often structural noise and external interference noise in the A-scan echo. The noise is similar to the defect echo in amplitude and position, which greatly interferes with the identification of defects. The noise reduction algorithm in the present invention learns the characteristics of defect signals during the neural network training process, so it can effectively identify useful signals in noisy A-scan echoes and remove interference signals. After the noise reduction algorithm of the present invention is used to reduce the noise of the A-scan signal, the structural noise and external interference noise can be effectively removed, which is convenient for the identification of defect echo signals.

[0043] Furthermore, in the preprocessing step of the ultrasonic A-scan signal, Gaussian noise of different decibels is added to the noise-free defect A-scan echo signal obtained by simulation to simulate various types of noise in the actual detection process. Using the signal after preprocessing in step S1 to construct a data set to train the neural network can enable the neural network to learn the characteristics of useful defect signals in various types of noise with different amplitudes, thereby enhancing the robustness and generalization ability of the model.

[0044] Furthermore, the convolution layer and the maximum pooling layer in the downsampling process of the U-Net denoising neural network can effectively extract features from the original signal, separate the useful features from the noise to the greatest extent, and then restore the signal after feature extraction to its original size through the transposed convolution layer in the upsampling process to achieve signal noise reduction. In addition, the jump connection layer in the U-Net neural network can splice the low-dimensional features with the high-dimensional features in the upsampling process, effectively retaining the low-dimensional useful features without losing them in the downsampling process.

[0045] Furthermore, the channel attention mechanism is used to perform global average pooling and global maximum pooling on the features for dimensionality reduction, and the results of average pooling and maximum pooling are obtained respectively, and then input into the shared multi-layer perceptron (MLP) for learning. The output of MLP is compressed by the sigmoid activation function to obtain the weight of each channel. The weight is used to enhance or weaken the contribution of the specified channel, weaken the weight of the noise feature, and increase the weight of the useful feature, so that the model can automatically learn the importance of each channel, more effectively utilize the input information of each channel, improve the performance of the model, and enhance the model denoising effect. The spatial attention mechanism can dynamically adjust the importance of different spatial positions of the input data in the neural network, enhance useful information and suppress useless information, improve the feature expression of key areas, and improve the representation ability and performance of the model. In addition, the spatial attention mechanism has strong adaptability and can adapt to most working environments and algorithm tasks.

[0046] Furthermore, the mean square error loss function is a widely used metric for evaluating prediction errors in regression problems. The mean square error loss function gives a greater penalty for larger errors because the square of the error will amplify the impact of large errors, which helps the model reduce large prediction deviations during training. And the mean square error is a differentiable function, which makes it possible to effectively find the minimum of the loss function through optimization algorithms such as gradient descent. It is a popular gradient descent optimization algorithm that combines the advantages of momentum and adaptive learning rate adjustment. The Adam optimizer is designed to deal with problems with non-stationary objectives, that is, when the optimal solution of the objective function changes during training, it can still converge quickly. And Adam is not very sensitive to the choice of learning rate. Even if the learning rate is not optimal, reasonable results can be obtained, which reduces the difficulty of adjusting parameters during neural network training.

[0047] It can be understood that the beneficial effects of the second aspect mentioned above can be found in the relevant description of the first aspect mentioned above, and will not be repeated here.

[0048] In summary, the present invention can improve the stability, robustness and universality of the denoising model, so that it can effectively denoise various types of noise.

[0049] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 It is a neural network structure diagram;

[0051] Figure 2 This is the structure diagram of the channel attention mechanism;

[0052] Figure 3 This is the structure diagram of the spatial attention mechanism;

[0053] Figure 4 This is the CABM module structure diagram;

[0054] Figure 5 This is a clean simulation signal diagram;

[0055] Figure 6 is a noisy signal graph;

[0056] Figure 7 is the noise reduction signal diagram;

[0057] Figure 8 A-scan signal with noise;

[0058] Fig. 9 Noise-reduced A-scan signal diagram;

[0059] Fig.10 Signal comparison before and after noise reduction;

[0060] Fig.11 A schematic diagram of a computer device provided by an embodiment of the present invention;

[0061] Fig.12 A block diagram of an electronic device provided according to an embodiment of the present invention;

[0062] Fig.13 It is a schematic diagram of the process of the present invention. DETAILED DESCRIPTION

[0063] 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 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.

[0064] In the description of the present invention, it should be understood that the terms “include” and “comprises” indicate the presence of described features, wholes, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or collections thereof.

[0065] It should also be understood that the terms used in the present specification are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the present specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0066] It should be further understood that the term "and / or" used in the present specification and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, and includes these combinations. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in the present invention generally indicates that the associated objects are in an "or" relationship.

[0067] It should be understood that, although the terms first, second, third, etc. may be used to describe preset ranges, etc. in the embodiments of the present invention, these preset ranges should not be limited to these terms. These terms are only used to distinguish preset ranges from each other. For example, without departing from the scope of the embodiments of the present invention, the first preset range may also be referred to as the second preset range, and similarly, the second preset range may also be referred to as the first preset range.

[0068] The word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrases "if it is determined" or "if (stated condition or event) is detected" may be interpreted as "when it is determined" or "in response to determining" or "when detecting (stated condition or event)" or "in response to detecting (stated condition or event)", depending on the context.

[0069] Various structural schematic diagrams of the embodiments disclosed in the present invention are shown in the accompanying drawings. These figures are not drawn to scale, and some details are magnified and some details may be omitted for the purpose of clear expression. The shapes of various regions and layers shown in the figures and the relative sizes and positional relationships therebetween are only exemplary, and may deviate in practice due to manufacturing tolerances or technical limitations, and those skilled in the art may additionally design regions / layers with different shapes, sizes, and relative positions according to actual needs.

[0070] The present invention provides an ultrasonic nondestructive testing A-scan signal denoising method based on an improved U-NET. The noise-free ultrasonic A-scan signal is preprocessed, including adding Gaussian noise of different decibels to the signal and normalizing it. The processed signal is then input into a constructed neural network structure, and weight parameters of each layer of the neural network are trained. The learning rate and the number of learning iterations are adjusted to improve the recognition accuracy of the neural network for useful features. After the network training is completed, the original signal with noise is input for denoising.

[0071] See also Fig.13 The present invention provides an ultrasonic nondestructive testing A-scan signal denoising method based on improved U-NET, comprising the following steps:

[0072] S1. Preprocessing the ultrasonic A-scan signal, including adding Gaussian noise of different decibels to the signal and normalizing it;

[0073] Since the U-Net-based denoising neural network designed in the present invention needs to learn how to recover the original signal from the data with added noise, it is necessary to input the noise-free signal and the corresponding noise-added signal at the same time so that the neural network can learn the effective features in the original signal; and in order to simulate various types of noise in real situations, it is necessary to add various decibels of noise to the A-scan signal to improve the robustness of the U-Net-based denoising neural network.

[0074] S2. Build a denoising neural network based on U-Net;

[0075] See also Figure 1 The structural scale of the U-Net-based denoising neural network is designed according to different application requirements and computing resources; since the data length of the A-scan signal is not long, the present invention constructs a U-Net-based denoising neural network consisting of an encoding part, a central part, a decoding part and an output layer.

[0076] In the encoding part, it consists of two convolutional layers, two batch normalization layers, and two maximum pooling layers;

[0077] The central part consists of two convolutional layers and two max pooling layers;

[0078] The decoding part consists of two transposed convolutional layers, two convolutional layers, and two batch normalization layers;

[0079] The output layer consists of a point-wise convolutional layer.

[0080] Without changing the feature dimension, the features are converted to the final number of output channels. At the same time, in order to prevent some high-dimensional features from being lost during the convolution process, a jump connection layer is used to fuse the feature maps of the downsampling process during the upsampling process.

[0081] The activation function between the convolution layer and the normalization layer is the PRelu activation function, which is expressed as follows:

[0082]

[0083] In the U-Net-based denoising neural network, the structures of each layer are as follows:

[0084] The convolution kernel space size in the encoding layer is 3*1 elements, and the stride is 1 element;

[0085] The spatial size of the maximum pooling layer kernel is 2*1 elements, and the stride is 2 elements;

[0086] The convolution kernel space size in the center layer is 3*1 element, and the stride is 1 element;

[0087] The spatial size of the transposed convolution kernel in the decoder layer is 2*1 elements, and the stride is 2 elements;

[0088] The spatial size of the convolution kernel is 3*1 element, and the stride is 1 element.

[0089] After passing through the convolution layer, the original one-dimensional signal can be transformed into multiple local features. The subsequent maximum pooling layer extracts the local features by reducing the dimension, reducing the number of parameters and the amount of calculation, preventing overfitting and improving network efficiency. After feature extraction and compression by the encoder layer, the signal is reconstructed by the decoder layer, and the useful features extracted by the encoding layer are reconstructed into a clean signal. At the same time, the use of jump connections between the encoding layer and the decoding layer allows low-level features to be passed to the final stage, avoiding the loss of some useful low-level features in the one-dimensional signal.

[0090] At the same time, the neural network model of the present invention also adds a channel attention mechanism and a spatial attention mechanism. The attention mechanism can make the model more focused when processing time series data, suppress irrelevant information, improve robustness and generalization ability, better capture useful features in the input data, and improve the model's resilience.

[0091] See also Figure 2 , the channel attention mechanism allows the model to automatically learn the importance of each channel, more effectively utilize the input information of each channel, and improve the performance of the model. The channel attention mechanism used in the present invention performs global average pooling and global maximum pooling on the feature map to reduce the dimension, obtains the results of average pooling and maximum pooling respectively, and then inputs them into a shared multi-layer perceptron (MLP) for learning. The output of the MLP is compressed by a sigmoid activation function to obtain the weight of each channel, and the contribution of the specified channel is enhanced or weakened by the weight.

[0092] See also Figure 3 ,The spatial attention mechanism in the neural network can dynamically adjust the ,importance of different spatial positions of the input data, enhance the useful ,information and suppress the useless information, improve the feature expression of the ,key areas, and improve the representation ability and performance of the ,model.

[0093] See also Figure 4 ,The present paper connects the channel attention mechanism and the spatial attention mechanism in series to form a CABM module, which is inserted into the neural network to improve the performance of the model.

[0094] The neural network structure of the present invention is composed of the above-mentioned parts and layers, which improves the feature extraction accuracy and the robustness and adversarial nature of the model compared to the traditional U-Net neural network.

[0095] S3, using the ultrasonic A-scan signal obtained in step S1 to train the U-Net-based denoising neural network obtained in step S2, and then inputting the original signal with noise into the trained neural network to achieve noise reduction processing.

[0096] The mean square error loss function and Adam optimizer are used to train the constructed neural network. The mean square error loss function is expressed as:

[0097]

[0098] Where N is the number of samples, x i is the predicted value, y i is the true value. The mean square error loss function is used to impose a greater penalty on larger error values, which can improve the accuracy of the model in predicting the true value.

[0099] Adam optimizer is an adaptive gradient descent optimizer, which can modify the descent direction and descent step size according to the descent gradient, thereby ensuring that the loss function can quickly decrease and converge during the forward training of the neural network.

[0100] The neural network model designed by the present invention needs to input a clean signal and add Gaussian noise of different decibels to simulate various types of noise in reality, and input the noisy signal into the model for training, and obtain the best noise reduction parameters by comparing with the clean signal to achieve noise reduction of the noise signal. And because different decibels of noise are added to simulate the actual situation, the noise reduction effect of the model on various types of noise in actual production is greatly improved, and the robustness of the model is improved.

[0101] The model training data set contains five groups of noisy data at 10dB, 15dB, 20dB, 25dB, and 30dB, a total of 200 groups of data, and 100 training rounds.

[0102] It will be appreciated by those skilled in the art that various aspects of the present invention may be implemented as systems, methods or program products. Therefore, various aspects of the present invention may be specifically implemented in the following forms, namely: complete hardware implementation, complete software implementation (including firmware, microcode, etc.), or a combination of hardware and software implementations, which may be collectively referred to herein as "circuits", "modules" or "platforms".

[0103] In another embodiment of the present invention, an ultrasonic nondestructive testing A-scan signal denoising system based on an improved U-NET is provided. The system can be used to implement the above-mentioned ultrasonic nondestructive testing A-scan signal denoising method based on the improved U-NET. Specifically, the ultrasonic nondestructive testing A-scan signal denoising system based on the improved U-NET includes a preprocessing module, a network module and a denoising module.

[0104] Wherein, the preprocessing module preprocesses the ultrasonic A-scan signal;

[0105] Network module, build a denoising neural network based on U-Net;

[0106] The denoising module uses the preprocessed ultrasonic A-scan signal to train the U-Net-based denoising neural network, and then inputs the noisy original signal into the trained neural network to achieve denoising processing.

[0107] In another embodiment of the present invention, a terminal device is provided, which includes a processor and a memory, wherein the memory is used to store a computer program, the computer program includes program instructions, and the processor is used to execute the program instructions stored in the computer storage medium. The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, which is suitable for implementing one or more instructions, and is specifically suitable for loading and executing one or more instructions to implement the corresponding method flow or corresponding function; the processor described in the embodiment of the present invention can be used for the operation of the ultrasonic non-destructive testing A-scan signal denoising method based on the improved U-NET, including:

[0108] Preprocess the ultrasonic A-scan signal; build a denoising neural network based on U-Net; use the obtained ultrasonic A-scan signal to train the denoising neural network based on U-Net, and then input the noisy original signal into the trained neural network to achieve noise reduction processing.

[0109] In another embodiment of the present invention, the present invention also provides a storage medium, specifically a computer-readable storage medium (Memory), which is a memory device in a terminal device for storing programs and data. It is understandable that the computer-readable storage medium here can include both the built-in storage medium in the terminal device and the extended storage medium supported by the terminal device, and can be any tangible medium containing or storing a program, which can be used by an instruction execution system, device or device or used in combination with it. The computer-readable storage medium provides a storage space, which stores the operating system of the terminal. In addition, one or more instructions suitable for being loaded and executed by the processor are also stored in the storage space, and these instructions can be one or more computer programs (including program codes). It should be noted that more specific examples (non-exhaustive list) of the computer-readable storage medium here include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0110] Computer readable storage media also include data signals propagated in baseband or as part of a carrier wave, which carry readable program codes. Such propagated data signals can take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The readable storage medium can also be any readable medium other than a readable storage medium, which can send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, device, or device. The program code contained on the readable storage medium can be transmitted using any appropriate medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination of the above.

[0111] Program code for performing the operations of the present invention may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., through the Internet using an Internet service provider).

[0112] The processor may load and execute one or more instructions stored in the computer-readable storage medium to implement the corresponding steps of the ultrasonic nondestructive testing A-scan signal noise reduction method based on the improved U-NET in the above embodiment; the processor may load and execute the following steps:

[0113] Preprocess the ultrasonic A-scan signal; build a denoising neural network based on U-Net; use the obtained ultrasonic A-scan signal to train the denoising neural network based on U-Net, and then input the noisy original signal into the trained neural network to achieve noise reduction processing.

[0114] See also Fig.11 , the terminal device is a computer device, and the computer device 60 of this embodiment includes: a processor 61, a memory 62, and a computer program 63 stored in the memory 62 and executable on the processor 61. When the computer program 63 is executed by the processor 61, the method for calculating the composition of fluid in the reservoir transformation wellbore in the embodiment is realized. To avoid repetition, it is not described one by one here. Alternatively, when the computer program 63 is executed by the processor 61, the functions of each model / unit in the calculation system for the composition of fluid in the reservoir transformation wellbore in the embodiment are realized. To avoid repetition, it is not described one by one here.

[0115] The computer device 60 may be a computing device such as a desktop computer, a notebook, a PDA, or a cloud server. The computer device 60 may include, but is not limited to, a processor 61 and a memory 62. Those skilled in the art will appreciate that Fig.11 This is only an example of the computer device 60 and does not constitute a limitation of the computer device 60. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the computer device may also include input and output devices, network access devices, buses, etc.

[0116] The processor 61 may be a central processing unit (CPU), or other general-purpose processors, central processing units, graphics processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, data processing logic based on quantum computing, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may be any conventional processor, etc.

[0117] The memory 62 may be an internal storage unit of the computer device 60, such as a hard disk or memory of the computer device 60. The memory 62 may also be an external storage device of the computer device 60, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc., equipped on the computer device 60.

[0118] Furthermore, the memory 62 may include both an internal storage unit of the computer device 60 and an external storage device. The memory 62 is used to store computer programs and other programs and data required by the computer device. The memory 62 may also be used to temporarily store data that has been output or is to be output.

[0119] Any reference to memory, database or other media used in the embodiments provided in the present application may include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory may include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM may be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0120] The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. The non-relational database may include a distributed database based on blockchain, etc., but is not limited thereto. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but is not limited thereto.

[0121] See also Fig.12 The terminal device 600 is an electronic device, and the electronic device is in the form of a general computing device. The components of the electronic device may include but are not limited to: at least one processing unit 610, at least one storage unit 620, a bus 630 connecting different platform components (including the storage unit 620 and the processing unit 610), a display unit 640, etc.

[0122] The storage unit stores program codes, which can be executed by the processing unit 610, so that the processing unit 610 performs the steps according to various exemplary embodiments of the present invention described in the above method section of this specification. For example, the processing unit 610 can perform the following steps: Fig.13 Follow the steps shown in .

[0123] The storage unit 620 may include a readable medium in the form of a volatile storage unit, such as a random access memory unit (RAM) 6201 and / or a cache memory unit 6202 , and may further include a read-only memory unit (ROM) 6203 .

[0124] The storage unit 620 may also include a program / utility 6204 having a set (at least one) of program modules 6205, such program modules 6205 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.

[0125] Bus 630 may represent one or more of several types of bus structures, including a memory unit bus or memory unit controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.

[0126] The electronic device 600 may also communicate with one or more external devices 700 (e.g., keyboards, pointing devices, Bluetooth devices, etc.), may also communicate with one or more devices that enable a user to interact with the electronic device 600, and / or communicate with any device that enables the electronic device 600 to communicate with one or more other computing devices (e.g., routers, modems, etc.). Such communication may be performed via an input / output (I / O) interface 650. Furthermore, the electronic device 600 may also communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) via a network adapter 660. The network adapter 660 may communicate with other modules of the electronic device 600 via a bus 630. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 600, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage platforms, etc.

[0127] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. The components of the embodiments of the present invention described and shown in the drawings here can usually be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the present invention. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0128] See also Figure 5 , Figure 6 and Figure 7 , is the noise reduction effect of this method on the simulated signal containing 20db Gaussian noise, from Figure 5 , Figure 6 and Figure 7 It can be seen that the present invention realizes the separation of the useful features of the original clean signal from the 20db Gaussian noise, and the defect echo position and amplitude are no different from the original clean signal, which verifies the feasibility of the neural network denoising scheme in the present invention.

[0129] See also Figure 8 , Fig. 9 and Fig.10 , which is a sample example and result diagram obtained after the model trained by the present invention is used to reduce the noise of the defect echo of the carbon fiber composite material in the actual production process, from Figure 8 , Fig. 9 and Fig.10 It can be seen that the present invention has a considerable effect of reducing the noise of the A-scan signal of carbon fiber composite material defect detection in the actual production process. Fig.10 From the comparison, it can be seen that the surface wave and defect echo have not changed in position compared with the original signal, and the amplitude has been reduced to a small extent, but the structural noise and external disturbance noise in the original signal have been greatly reduced, which greatly improves the accuracy of distinguishing defect echoes from A-scan signals and facilitates subsequent three-dimensional reconstruction.

[0130] In order to verify the superiority of the present invention over the traditional denoising model, the original signal containing noise with the same signal-to-noise ratio is denoised using the traditional U-Net neural network and the autoencoder neural network respectively and compared with the present invention. In order to ensure the objectivity and stability of the data, the denoising effect of each model for different decibel noises is tested respectively. The final results are shown in Table 1.

[0131] Table 1

[0132]

[0133] It can be seen from Table 1 that the noise reduction model designed in the present invention has better noise reduction effects for each decibel of noise than other models, and the present invention has very good practicality.

[0134] In summary, the ultrasonic nondestructive testing A-scan signal denoising method and system based on improved U-NET of the present invention has the following advantages:

[0135] 1) Based on the traditional U-Net neural network, the channel attention mechanism and spatial attention mechanism are introduced to improve the model's accuracy in extracting useful features.

[0136] 2) For preliminary non-destructive testing such as the presence or absence of defects based on A-scan signals, the present invention can greatly reduce the interference of environmental noise on the detection signal and improve the signal-to-noise ratio, thereby reducing the professional and technical requirements for engineers and the influence of subjective factors in the processing process.

[0137] 3) Based on the traditional U-Net neural network, the channel attention mechanism and spatial attention mechanism are introduced to improve the model's accuracy in extracting useful features.

[0138] 4) For anisotropic test objects such as carbon fiber composite materials, the present invention greatly reduces the interference of structural noise, improves the detection rate of defects and the accuracy of defect positioning, and lays a foundation for subsequent three-dimensional reconstruction, thereby improving the accuracy of three-dimensional reconstruction defects.

[0139] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0140] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0141] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed in the present invention can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0142] In the embodiments provided by the present invention, it should be understood that the disclosed devices / terminals and methods can be implemented in other ways. For example, the device / terminal embodiments described above are only schematic. For example, the division of the modules or units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0143] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0144] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0145] If the integrated module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0146] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0147] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0148] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0149] The above contents are only for explaining the technical idea of ​​the present invention and cannot be used to limit the protection scope of the present invention. Any changes made on the basis of the technical solution in accordance with the technical idea proposed by the present invention shall fall within the protection scope of the claims of the present invention.

Claims

1. A method for reducing noise of ultrasonic nondestructive testing A-scan signals based on improved U-NET, characterized in that: The following steps are involved: S1, preprocessing the ultrasonic A-scan signal; S2. Build a denoising neural network based on U-Net; S3, using the ultrasonic A-scan signal obtained in step S1 to train the U-Net-based denoising neural network obtained in step S2, and then inputting the original signal with noise into the trained neural network to achieve noise reduction processing.

2. The method for reducing noise of ultrasonic nondestructive testing A-scan signals based on improved U-NET according to claim 1, characterized in that: Step S1 is specifically as follows: Gaussian noise of different decibels is added to the noise-free ultrasound A-scan signal and then normalized.

3. The method for reducing noise of ultrasonic nondestructive testing A-scan signals based on improved U-NET according to claim 1, characterized in that: In step S2, the U-Net-based denoising neural network includes an encoding part, a central part, a decoding part, and an output layer. The features are converted into the final number of output channels without changing the feature dimension, and a skip connection layer is used to fuse the feature maps in the downsampling process in the upsampling process; The encoding part includes two convolutional layers, two batch normalization layers, and two maximum pooling layers; The central part includes two convolutional layers and two maximum pooling layers; The decoding part includes two transposed convolutional layers, two convolutional layers, and two batch normalization layers; The output layer consists of a point-wise convolutional layer.

4. The method for reducing noise of ultrasonic nondestructive testing A-scan signals based on improved U-NET according to claim 3, characterized in that: The activation function between the convolution layer and the normalization layer is the PRelu activation function as follows: Among them, α is a learnable parameter and x is the input of the activation function.

5. The method for reducing noise of ultrasonic nondestructive testing A-scan signals based on improved U-NET according to claim 3, characterized in that: The convolution kernel space size in the encoding layer is 3*1 elements, and the stride is 1 element; The spatial size of the maximum pooling layer kernel is 2*1 elements, and the stride is 2 elements; The convolution kernel space size in the center layer is 3*1 element, and the stride is 1 element; The spatial size of the transposed convolution kernel in the decoder layer is 2*1 elements, and the stride is 2 elements; The spatial size of the convolution kernel is 3*1 element, and the stride is 1 element.

6. The method for reducing noise of ultrasonic nondestructive testing A-scan signals based on improved U-NET according to claim 3, characterized in that: The U-Net-based denoising neural network also includes a channel attention mechanism and a spatial attention mechanism; the channel attention mechanism and the spatial attention mechanism are connected in series to form a CABM module, which is then inserted into the U-Net-based denoising neural network.

7. The method for reducing noise of ultrasonic nondestructive testing A-scan signals based on improved U-NET according to claim 6, characterized in that: The channel attention mechanism performs global average pooling and global maximum pooling on the feature map to reduce the dimension, obtains the results of average pooling and maximum pooling respectively, and then inputs them into the shared multi-layer perceptron for learning. The output of the multi-layer perceptron is compressed by the sigmoid activation function to obtain the weight of each channel, and the weight is used to enhance or weaken the contribution of the specified channel. The spatial attention mechanism dynamically adjusts the importance of different spatial positions of the input data in the U-Net based denoising neural network.

8. The method for reducing noise of ultrasonic nondestructive testing A-scan signals based on improved U-NET according to claim 1, characterized in that: In step S3, the mean square error loss function and the Adam optimizer are used to train the U-Net-based denoising neural network.

9. The method for reducing noise of ultrasonic nondestructive testing A-scan signals based on improved U-NET according to claim 8, characterized in that: The mean square error loss function MSE is: Where N is the number of samples, x i is the predicted value, y i is the true value.

10. An ultrasonic nondestructive testing A-scan signal noise reduction system based on improved U-NET, characterized in that: include: A preprocessing module, which preprocesses the ultrasonic A-scan signal; Network module, build a denoising neural network based on U-Net; The denoising module uses the preprocessed ultrasonic A-scan signal to train the U-Net-based denoising neural network, and then inputs the noisy original signal into the trained neural network to achieve denoising processing.

Images