Defect identification method and device for power transmission crimping fitting

Abstract

The embodiment of the invention provides a defect identification method and device for a power transmission crimping fitting. The method comprises the following steps: acquiring an acoustic signal of the power transmission crimping fitting; performing noise reduction processing on the acoustic signal to obtain a noise-reduced acoustic signal; performing feature processing on the acoustic signal after noise reduction to obtain features after feature processing; performing three-dimensional imaging reconstruction on the features after feature processing to obtain a three-dimensional acoustic image; and performing defect identification on the three-dimensional acoustic image to obtain a defect type. According to the embodiment of the invention, the noise reduction processing and the feature processing are carried out on the acquired acoustic signal to obtain the feature after the feature processing, the three-dimensional imaging reconstruction is carried out on the feature after the feature processing to obtain the three-dimensional acoustic image, and the defect identification is carried out based on the three-dimensional acoustic image, so that the high-precision detection of the power transmission crimping fitting can be realized; and the accuracy and reliability of defect identification are improved.

Description

Technical Field

[0001] The present application relates to the field of detection technology, and in particular to a method and device for identifying defects in power transmission connection fittings. Background Art

[0002] Currently, there are several main methods for identifying defects in power transmission connectors: visual inspection, resistance measurement, and traditional ultrasonic testing. However, visual inspection can only detect visible defects on the surface of power transmission connectors and cannot deeply examine the internal structure. Resistance measurement focuses on evaluating electrical conductivity and is difficult to effectively identify internal defects. While traditional ultrasonic testing can perform internal inspections, it presents results in two-dimensional planar images, making defect identification particularly difficult for complex power transmission connectors and resulting in low accuracy. Summary of the Invention

[0003] The present application provides a method and device for identifying defects in power transmission connection fittings, which can improve the accuracy and reliability of defect identification.

[0004] In a first aspect, an embodiment of the present application provides a method for identifying defects in power transmission connecting fittings, comprising: collecting acoustic signals of the power transmission connecting fittings; performing noise reduction processing on the acoustic signals to obtain noise-reduced acoustic signals; performing feature processing on the noise-reduced acoustic signals to obtain features after feature processing; performing three-dimensional imaging reconstruction on the features after feature processing to obtain a three-dimensional acoustic image; and performing defect identification on the three-dimensional acoustic image to obtain a defect type.

[0005] In the second aspect, an embodiment of the present application also provides a defect identification device for power transmission connecting fittings, including: an acoustic signal acquisition module for collecting the acoustic signal of the power transmission connecting fittings; a noise reduction processing module for performing noise reduction processing on the acoustic signal to obtain the acoustic signal after noise reduction; a feature processing module for performing feature processing on the acoustic signal after noise reduction to obtain the feature after feature processing; a three-dimensional imaging reconstruction module for performing three-dimensional imaging reconstruction on the feature after feature processing to obtain a three-dimensional acoustic image; and a defect identification module for performing defect identification on the three-dimensional acoustic image to obtain the defect type.

[0006] In a third aspect, an embodiment of the present application further provides an electronic device, comprising: one or more processors; a storage device for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the defect identification method for voltage transmission connection fittings as described in the embodiment of the present application.

[0007] In a fourth aspect, an embodiment of the present application further provides a storage medium comprising computer-executable instructions, which, when executed by a computer processor, are used to execute the defect identification method for power transmission connection fittings as described in an embodiment of the present application.

[0008] In a fifth aspect, an embodiment of the present application further provides a computer program product, including a computer program, which, when executed by a processor, implements the defect identification method for power transmission connection fittings as described in the embodiment of the present application.

[0009] The technical solution of the embodiment of the present application collects the acoustic signal of the power transmission connection fittings; performs noise reduction processing on the acoustic signal to obtain the noise-reduced acoustic signal; performs feature processing on the noise-reduced acoustic signal to obtain the feature after feature processing; performs three-dimensional imaging reconstruction on the feature after feature processing to obtain a three-dimensional acoustic image; and performs defect identification on the three-dimensional acoustic image to obtain the defect type. The embodiment of the present application performs noise reduction processing and feature processing on the collected acoustic signal to obtain the feature after feature processing, and performs three-dimensional imaging reconstruction on the feature after feature processing to obtain a three-dimensional acoustic image, and performs defect identification based on the three-dimensional acoustic image, which can achieve high-precision detection of power transmission connection fittings, improve the accuracy and reliability of defect identification, and meet the actual needs of efficient operation and maintenance of power equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] The above and other features, advantages, and aspects of the various embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and that the originals and elements are not necessarily drawn to scale.

[0011] Figure 1 A schematic flow chart of a method for identifying defects in power transmission connection fittings provided in an embodiment of the present application;

[0012] Figure 2 A schematic flow chart of another method for identifying defects in power transmission connection fittings provided in an embodiment of the present application;

[0013] Figure 3 A schematic structural diagram of a defect identification device for a power transmission connection fitting provided in an embodiment of the present application;

[0014] Figure 4 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0015] The following describes embodiments of the present disclosure in more detail with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as limited to the embodiments described herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are for illustrative purposes only and are not intended to limit the scope of protection of the present disclosure.

[0016] It should be understood that the various steps described in the method implementation of the present disclosure can be performed in different orders and / or in parallel. In addition, the method implementation may include additional steps and / or omit the steps shown. The scope of the present disclosure is not limited in this respect. The term "including" and its variations used herein are open inclusions, that is, "including but not limited to". It should be noted that the concepts of "first", "second", etc. mentioned in this disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units. It should be noted that the modifications of "one" and "multiple" mentioned in this disclosure are illustrative and not restrictive. Those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more". It is understandable that the data involved in this technical solution (including but not limited to the data itself, the acquisition or use of the data) should comply with the requirements of relevant laws, regulations and relevant provisions.

[0017] Figure 1 This is a flow chart of a method for identifying defects in power transmission fittings provided by an embodiment of the present application. This embodiment of the present application is applicable to situations where the types of defects in power transmission fittings are detected. The method can be executed by a device for identifying defects in power transmission fittings, which can be implemented in the form of software and / or hardware. Optionally, it can be implemented by an electronic device, which can be a mobile terminal, PC, or server. Figure 1 As shown, the method includes:

[0018] S110: Collect acoustic signals of the voltage transmission connection fittings.

[0019] Among them, transmission voltage connecting fittings are core components used to ensure the electrical connection and mechanical fixation of transmission lines in the power transmission system. Their performance is directly related to the safety and stability of power grid operation.

[0020] In this embodiment, a transmitting device may transmit a continuous ultrasonic signal to the power transmission connection fitting, and a collecting device may collect signals reflected and transmitted by the ultrasonic signal, which are recorded as acoustic signals.

[0021] Optionally, collecting acoustic signals of the power transmission connection fittings includes: transmitting continuous ultrasonic signals to the power transmission connection fittings according to adjustment parameters of the transmitting device; wherein the adjustment parameters include at least one of the following: operating frequency, array element excitation timing, and emission angle; collecting multiple acoustic signals through the collection device and according to the set sampling frequency; wherein the acoustic signals are reflected wave signals and transmitted wave signals corresponding to the ultrasonic signals; and the spatial position information and depth information inside the power transmission connection fitting corresponding to each acoustic signal are different.

[0022] In this embodiment, there are no restrictions on the transmitting device and the collecting device. For example, the transmitting device can be a laser ultrasonic device, and the corresponding collecting device can be a laser interferometer device. Both the transmitting device and the corresponding collecting device can be electromagnetic ultrasonic devices. Laser ultrasonic devices use high-energy pulsed lasers to irradiate the surface of the hardware, causing localized surface heating to generate ultrasonic waves, and then receive the reflected wave signals through the laser interferometer device. This method does not require a coupling agent and is suitable for detection in harsh environments such as high temperature and high pressure. However, the equipment is expensive, the stability and accuracy requirements of the laser system are extremely high, and the collected signals are easily affected by fluctuations in laser energy. As for electromagnetic ultrasonic devices, they are based on the principles of electromagnetic induction and Lorentz force and can excite and receive ultrasonic signals without direct contact with the power supply connection hardware. They can be used on power supply connection hardware with complex shapes and rough surfaces, but the detection sensitivity is relatively low, the signal strength is weak, and they are significantly affected by electromagnetic interference, requiring additional shielding measures.

[0023] Optionally, the transmitting device is a phased array acoustic probe; the surface of the transmitting device and the surface of the voltage transmission contact fitting include a set coupling agent.

[0024] A phased array acoustic probe is an acoustic detection device that can adjust the operating frequency (for example, the operating frequency can be adjusted within a range of 2MHz-10MHz), the timing of array element excitation, and the angle of ultrasonic signal emission. By controlling the ultrasonic emission time difference between each array element of the phased array acoustic probe, ultrasonic focusing and scanning are achieved, thereby acquiring acoustic signals at different angles and depths.

[0025] The coupling agent is a material composed of a high-viscosity elastic colloid and an acoustic impedance matching agent. Its acoustic impedance is highly matched to the materials of the power transmission connector and the phased array acoustic probe. When applied between the phased array acoustic probe and the test surface of the power transmission connector, it effectively reduces acoustic wave reflection losses at the interface, enhances acoustic wave coupling efficiency, and ensures that the transmitted ultrasonic signal can be smoothly transmitted into the power transmission connector for testing.

[0026] Among them, the collection device is an acoustic sensor array; the acoustic sensor array is composed of multiple acoustic sensors, which are distributed around the power transmission connection fittings according to a set shape.

[0027] The acoustic sensor array is a distributed acoustic sensor array consisting of multiple highly sensitive acoustic sensors arranged in a predetermined shape (e.g., a circle). This array can simultaneously collect reflected and transmitted wave signals at different locations around the power transmission connector, generating multi-channel, multi-dimensional acoustic signals and providing rich data for subsequent analysis.

[0028] In this embodiment, a parameter-adjustable transmitting device (such as a phased array acoustic probe) can be used to dynamically adjust the probe's operating frequency (adjustment range is 2-10MHz), array element excitation timing, and acoustic wave emission angle based on the material properties (such as metal type, hardness, etc.), geometric shape (such as size specifications, structural characteristics), and detection accuracy requirements of the voltage input connection hardware. A special coupling agent is evenly applied between the probe and the hardware detection surface. The coupling agent is composed of a high-viscosity elastic colloid and an acoustic impedance matching agent. Its acoustic impedance is highly matched with the hardware and probe materials, which can effectively reduce the reflection loss of the sound wave at the interface and enhance the acoustic wave coupling efficiency. At the same time, a distributed acoustic sensor array is deployed. The array consists of a plurality of high-sensitivity acoustic sensors, which are distributed around the hardware according to a preset spatial layout (such as a circular shape). The reflected wave signals and transmitted wave signals of the hardware at different spatial positions and different depths are synchronously collected at a set sampling frequency (for example, 100kHz) to form a multi-channel, multi-dimensional acoustic signal.

[0029] For example, for a certain type of aluminum alloy power transmission fittings, a phased array acoustic probe with an operating frequency range of 3-8MHz is selected. According to the size specifications (such as length 150mm, width 80mm, thickness 20mm) and structural characteristics of the power transmission fittings, the probe array element excitation angle is adjusted to 45° through the probe control system, and the operating frequency is set to 5MHz. A special coupling agent is evenly applied to the detection surface of the probe and the power transmission fitting to ensure that the thickness of the coupling layer is uniform. At the same time, a distributed acoustic sensor array consisting of 16 acoustic sensors is deployed around the power transmission fittings. They are arranged in a square grid with a side length of 10mm, and the reflected wave signals and transmitted wave signals of the power transmission fittings at different positions and depths are synchronously collected at a set sampling frequency of 100kHz. The collection time is 6 seconds, and multi-channel and multi-dimensional acoustic signals are obtained.

[0030] S120: Perform noise reduction processing on the acoustic signal to obtain a noise-reduced acoustic signal.

[0031] In this embodiment, there is no limitation on the noise reduction method, and the noise reduction method may be, for example, adaptive filtering, wavelet threshold noise reduction, variational mode decomposition, etc.

[0032] Optionally, the acoustic signal is subjected to noise reduction processing to obtain a noise-reduced acoustic signal, including: decomposing the acoustic signal into multiple submodal components; wherein the frequency characteristics of the multiple submodal components are different; and noise reduction is performed on each of the multiple submodal components based on local statistical information of the signal and filtering parameters to obtain a noise-reduced acoustic signal.

[0033] The local statistical information of the signal may be a variance or a mean, etc. The filtering parameter may be a filtering threshold, an iteration step, etc.

[0034] In this embodiment, the acoustic signal can be decomposed into multiple submodal components by any noise reduction processing method, such as variational mode decomposition (VMD), to achieve effective frequency domain separation of the signal. Exemplarily, the number of submodal components can be 8. The frequency characteristics of each submodal component in the multiple submodal components are different. Then, adaptive sparse filtering is applied to each submodal component, and the filtering parameters are dynamically adjusted according to the local statistical information of the signal to remove noise, thereby retaining the effective acoustic signal characteristics to the greatest extent and significantly improving the signal-to-noise ratio of the signal.

[0035] S130: Perform feature processing on the acoustic signal after noise reduction to obtain features after feature processing.

[0036] In this embodiment, there is no limitation on the feature processing method. For example, it may include extracting multiple features such as time domain features and frequency domain features, and then fusing the extracted multiple features to obtain processed features. The fusion method can be deep belief network fusion or principal component analysis fusion.

[0037] Optionally, feature processing is performed on the acoustic signal after noise reduction to obtain features after feature processing, including: extracting the time domain features, frequency domain features and time-frequency domain features of the reflected wave signal respectively, and correspondingly obtaining the first modal feature vector, the second modal feature vector and the third modal feature vector; extracting the time domain features, frequency domain features and time-frequency domain features of the transmitted wave signal respectively, and correspondingly obtaining the fourth modal feature vector, the fifth modal feature vector and the sixth modal feature vector; extracting the first modal feature vector, the second modal feature vector, the third modal feature vector, the fourth modal feature vector, the fifth modal feature vector and the sixth modal feature vector Normalization processing is performed to obtain the normalized first modal eigenvector, the normalized second modal eigenvector, the normalized third modal eigenvector, the normalized fourth modal eigenvector, the normalized fifth modal eigenvector and the normalized sixth modal eigenvector; the normalized first modal eigenvector, the normalized second modal eigenvector, the normalized third modal eigenvector, the normalized fourth modal eigenvector, the normalized fifth modal eigenvector and the normalized sixth modal eigenvector are subjected to feature fusion through a deep belief network to obtain features after feature processing.

[0038] Among them, the deep belief network can be a deep learning model composed of multiple layers of restricted Boltzmann machines, which is used to perform unsupervised learning on the extracted modal feature vectors. By layer-by-layer training, the complex relationship between feature vectors is mined, and the fused feature vectors (i.e., the features after feature processing) are output to enhance the characterization ability of the internal structure information of the power transmission connection fittings.

[0039] The acoustic signal after noise reduction includes a reflected wave signal and a transmitted wave signal.

[0040] In this embodiment, time domain features (such as peak value, mean value, and variance), frequency domain features (such as spectral distribution and dominant frequency), and time-frequency domain features (such as time-frequency matrix features obtained using short-time Fourier transform) can be extracted from the de-noised reflected wave signal and the de-noised transmitted wave signal, respectively, to obtain corresponding eigenvectors. Extracting the time domain, frequency domain, and time-frequency domain features of the reflected wave signal can correspondingly obtain a first modal eigenvector, a second modal eigenvector, and a third modal eigenvector; extracting the time domain, frequency domain, and time-frequency domain features of the transmitted wave signal can correspondingly obtain a fourth modal eigenvector, a fifth modal eigenvector, and a sixth modal eigenvector. The first modal eigenvector, the second modal eigenvector, the third modal eigenvector, the fourth modal eigenvector, the fifth modal eigenvector, and the sixth modal eigenvector are normalized to obtain the normalized first modal eigenvector, the normalized second modal eigenvector, the normalized third modal eigenvector, the normalized fourth modal eigenvector, the normalized fifth modal eigenvector, and the normalized sixth modal eigenvector. In this embodiment, there is no limitation on the normalization method, and for example, minimum and maximum scaling and standard deviation normalization can be used.

[0041] In this embodiment, the normalized first modal feature vector, the normalized second modal feature vector, the normalized third modal feature vector, the normalized fourth modal feature vector, the normalized fifth modal feature vector, and the normalized sixth modal feature vector can be input into a deep belief network composed of a multi-layer (such as 3-layer) restricted Boltzmann machine. Through unsupervised learning, the potential correlation between different modal feature vectors is automatically learned and mined to achieve deep fusion of multimodal feature vectors, and the fused feature vector, that is, the feature after feature processing, is output.

[0042] S140 , performing three-dimensional imaging reconstruction on the features after feature processing to obtain a three-dimensional acoustic image.

[0043] In this embodiment, the processed features (two-dimensional information) may be converted into a three-dimensional acoustic image using a three-dimensional tomography algorithm.

[0044] S150: Perform defect recognition on the three-dimensional acoustic image to obtain the defect type.

[0045] In this embodiment, any deep learning algorithm can be used to identify defects in the 3D acoustic image to obtain the defect type. Exemplarily, the deep learning algorithm can be a convolutional neural network, an attention mechanism, or the like.

[0046] In this embodiment, there is no limitation on the defect type, and the defect types may be, for example, cracks, holes, looseness, and other different defect types.

[0047] The technical solution of the embodiment of the present application collects the acoustic signal of the power transmission connection fittings; performs noise reduction processing on the acoustic signal to obtain the noise-reduced acoustic signal; performs feature processing on the noise-reduced acoustic signal to obtain the feature after feature processing; performs three-dimensional imaging reconstruction on the feature after feature processing to obtain a three-dimensional acoustic image; and performs defect identification on the three-dimensional acoustic image to obtain the defect type. The embodiment of the present application performs noise reduction processing and feature processing on the collected acoustic signal to obtain the feature after feature processing, and performs three-dimensional imaging reconstruction on the feature after feature processing to obtain a three-dimensional acoustic image, and performs defect identification based on the three-dimensional acoustic image, which can achieve high-precision detection of power transmission connection fittings, improve the accuracy and reliability of defect identification, and meet the actual needs of efficient operation and maintenance of power equipment.

[0048] Figure 2 This is a flow chart of another method for identifying defects in power transmission connection fittings provided by the embodiment of the present application. This embodiment of the present application is a specific implementation of the above invention embodiment. Figure 2 The method provided in the embodiment of the present application specifically includes the following steps:

[0049] S201. Collect acoustic signals of voltage transmission connection fittings.

[0050] S202: Perform noise reduction processing on the acoustic signal to obtain a noise-reduced acoustic signal.

[0051] S203: Perform feature processing on the acoustic signal after noise reduction to obtain features after feature processing.

[0052] S204: Divide the internal space of the voltage transmission connection fitting into a plurality of voxel units.

[0053] The voxel unit may be the smallest volume unit in a three-dimensional space, similar to a pixel in a two-dimensional image.

[0054] In this embodiment, the internal space of the voltage transmission connection fitting is divided into tiny voxel units (eg, 5 mm×5 mm×5 mm) with a specific spatial resolution.

[0055] S205 : Determine initial acoustic parameters of each voxel unit in the plurality of voxel units according to the propagation information of the acoustic signal and the set sound wave propagation physical model.

[0056] In this embodiment, the initial acoustic parameters (including acoustic impedance, sound speed, attenuation coefficient, etc.) of each voxel unit can be calculated based on the propagation information of the acoustic signal (such as propagation time, amplitude, etc.) combined with the set sound wave propagation physical model.

[0057] S206: Input the initial acoustic parameters and the features after feature processing into a pre-trained deep learning image reconstruction network to obtain the target acoustic parameters of each voxel unit.

[0058] In this embodiment, a deep learning-based image reconstruction network can be used, with initial acoustic parameters and features after feature processing as input. Through the network training optimization process (supervised training is performed using a large amount of acoustic signal data of voltage-transmitting connectors with known internal structures), the initial acoustic parameters of each voxel unit are predicted and corrected to obtain the target acoustic parameters of each voxel unit.

[0059] S207 : Map the target acoustic parameters of each voxel unit according to the three-dimensional spatial position of each voxel unit to generate a three-dimensional acoustic image.

[0060] In this embodiment, the target acoustic parameters of each voxel unit are mapped layer by layer to the three-dimensional spatial position of each voxel unit (by stacking the voxels), reconstructing a high-precision three-dimensional acoustic image of the power transmission connection hardware. The three-dimensional acoustic image can present the internal structural information of the power transmission connection hardware in a three-dimensional visual form, intuitively showing the spatial distribution and morphological characteristics of internal defects in the object.

[0061] S208: Perform defect recognition on the three-dimensional acoustic image to obtain the defect type.

[0062] Optionally, performing defect recognition on the three-dimensional acoustic image to obtain the defect type includes: dividing the three-dimensional acoustic image into two-dimensional slice images; performing defect recognition on the two-dimensional slice images based on a set defect recognition model to obtain the defect type; wherein the defect recognition model is set to a convolutional neural network model.

[0063] In this embodiment, a 3D acoustic image is sliced ​​into 2D slice images by layer and fed into a predefined defect recognition model, such as a convolutional neural network model. For example, the convolutional neural network model may include multiple convolutional layers, pooling layers, and fully connected layers. The convolutional kernels of the convolutional layers extract local image features, the pooling layers reduce feature dimensionality, and the fully connected layers perform feature classification. After training with a large amount of labeled defect sample data, the convolutional neural network model can accurately identify different defect types, such as cracks, holes, and porosity.

[0064] This embodiment, based on a deep learning-based three-dimensional imaging reconstruction method and a set defect recognition model, breaks through the limitations of traditional two-dimensional imaging, achieving high-precision three-dimensional acoustic imaging of power transmission connection hardware and accurate identification and classification of defect types. It can intuitively and comprehensively present the three-dimensional morphology and distribution of defects within the hardware, providing strong support for defect analysis and evaluation.

[0065] S209: Annotate the three-dimensional acoustic image according to the annotation information and generate a defect analysis report.

[0066] In this embodiment, there is no specific restriction on the standard information. The annotation information includes at least the defect type. In addition to the defect type, it can also include the three-dimensional spatial coordinates, geometric dimensions, and severity assessment level of the defect to generate a defect analysis report.

[0067] S210 : Transmitting the three-dimensional acoustic image and the defect analysis report to a remote server, so that the remote server analyzes the three-dimensional acoustic image based on the defect analysis report.

[0068] In this embodiment, 3D acoustic images and defect analysis reports, among other inspection data, can be encrypted and transmitted to a remote server via a wireless network communication module (e.g., a 5G communication module). Dedicated visualization analysis software is configured on the remote server, allowing operators to perform multi-angle rotation, scaling, and sectioning of the 3D acoustic images, enabling comprehensive observation and analysis of the internal structure and defects of the power transmission connection fittings. This analysis allows the operator to formulate a scientific and rational equipment maintenance and repair strategy based on the results.

[0069] In this embodiment, detection data such as three-dimensional acoustic images and defect analysis reports can be remotely transmitted and visualized for analysis, making it easier for power operation and maintenance personnel to inspect and evaluate hardware at different locations, thereby realizing intelligent operation and maintenance management of power equipment, improving operation and maintenance efficiency, reducing operation and maintenance costs, and significantly improving the safety and reliability of power system operation.

[0070] This embodiment utilizes a collection method that combines a parameter-adjustable phased array acoustic probe with a distributed acoustic sensor array. This allows for comprehensive, multi-dimensional acoustic signal acquisition of various types of power supply connectors, acquiring rich internal structural information and providing a data foundation for high-precision imaging. Based on the acquired acoustic signals, an improved three-dimensional imaging reconstruction and deep learning image reconstruction network are employed to divide the internal space of the power supply connector into tiny voxel units. By calculating the acoustic parameters of each voxel unit, a high-precision three-dimensional acoustic image is reconstructed layer by layer. This intuitively displays the three-dimensional morphology and distribution of defects, breaking through the limitations of traditional two-dimensional imaging.

[0071] This embodiment uses a combined noise reduction technique of variational mode decomposition (VMD) and adaptive sparse filtering to decompose the acoustic signal into multiple submodal components based on the signal's frequency domain characteristics. The filter parameters for each component are then dynamically adjusted to remove noise, effectively suppressing interference from complex environmental noise. Simultaneously, a deep belief network composed of restricted Boltzmann machines is constructed to extract time, frequency, and time-frequency domain features from the denoised reflected and transmitted wave signals. The network automatically learns the potential correlations between different modal feature vectors, achieving deep fusion and significantly improving signal quality and feature extraction capabilities, enhancing anti-interference performance and detection accuracy in complex environments.

[0072] Figure 3 This is a schematic diagram of a defect identification device for a power transmission connection fitting provided in an embodiment of the present application, as shown in FIG. Figure 3 As shown, the device includes: an acoustic signal acquisition module 310, a noise reduction processing module 320, a feature processing module 330, a three-dimensional imaging reconstruction module 340 and a defect recognition module 350;

[0073] An acoustic signal acquisition module 310 is used to acquire the acoustic signal of the voltage transmission connection fitting;

[0074] The noise reduction processing module 320 is used to perform noise reduction processing on the acoustic signal to obtain a noise-reduced acoustic signal;

[0075] A feature processing module 330 is used to perform feature processing on the noise-reduced acoustic signal to obtain features after feature processing;

[0076] A three-dimensional imaging reconstruction module 340 is used to perform three-dimensional imaging reconstruction on the features after the feature processing to obtain a three-dimensional acoustic image;

[0077] The defect recognition module 350 is configured to perform defect recognition on the three-dimensional acoustic image to obtain a defect type.

[0078] The technical solution of the embodiment of the present application is to collect the acoustic signal of the power transmission connection fitting through the acoustic signal acquisition module; to perform noise reduction processing on the acoustic signal through the noise reduction processing module to obtain the acoustic signal after noise reduction; to perform feature processing on the acoustic signal after noise reduction through the feature processing module to obtain the feature after feature processing; to perform three-dimensional imaging reconstruction on the feature after feature processing through the three-dimensional imaging reconstruction module to obtain a three-dimensional acoustic image; and to perform defect recognition on the three-dimensional acoustic image through the defect recognition module to obtain the defect type. The embodiment of the present application performs noise reduction processing and feature processing on the collected acoustic signal to obtain the feature after feature processing, and performs three-dimensional imaging reconstruction on the feature after feature processing to obtain a three-dimensional acoustic image, and performs defect recognition based on the three-dimensional acoustic image, thereby achieving high-precision detection of power transmission connection fittings and improving the accuracy and reliability of defect recognition.

[0079] The defect identification device for power transmission connection fittings provided in the embodiments of the present application can execute the defect identification method for power transmission connection fittings provided in any embodiment of the present disclosure, and has the corresponding functional modules and beneficial effects of the execution method.

[0080] Optionally, the acoustic signal acquisition module is specifically used to: transmit continuous ultrasonic signals to the power transmission connection fitting according to the adjustment parameters of the transmitting device; wherein the adjustment parameters include at least one of the following: operating frequency, array element excitation timing and emission angle; collect multiple acoustic signals through the acquisition device and according to the set sampling frequency; wherein the acoustic signals are the reflected wave signals and the transmitted wave signals corresponding to the ultrasonic signals; the spatial position information and depth information inside the power transmission connection fitting corresponding to each acoustic signal are different.

[0081] Wherein, the transmitting device is a phased array acoustic probe; the surface of the transmitting device and the surface of the voltage transmission connection fitting include a set coupling agent.

[0082] Wherein, the acquisition device is an acoustic sensor array; the acoustic sensor array is composed of a plurality of acoustic sensors, which are distributed around the voltage transmission connection fitting according to a set shape.

[0083] Optionally, the noise reduction processing module is specifically used to: decompose the acoustic signal into multiple sub-modal components; wherein the frequency characteristics of the multiple sub-modal components are different; and perform noise reduction on each of the multiple sub-modal components based on local statistical information of the signal and filtering parameters to obtain a noise-reduced acoustic signal.

[0084] The acoustic signal after noise reduction includes a reflected wave signal and a transmitted wave signal.

[0085] Optionally, the feature processing module is specifically used to: extract the time domain features, frequency domain features and time-frequency domain features of the reflected wave signal respectively, and obtain the first modal feature vector, the second modal feature vector and the third modal feature vector respectively; extract the time domain features, frequency domain features and time-frequency domain features of the transmitted wave signal respectively, and obtain the fourth modal feature vector, the fifth modal feature vector and the sixth modal feature vector respectively; and perform normalization processing on the first modal feature vector, the second modal feature vector, the third modal feature vector, the fourth modal feature vector, the fifth modal feature vector and the sixth modal feature vector. Correspondingly, a normalized first modal feature vector, a normalized second modal feature vector, a normalized third modal feature vector, a normalized fourth modal feature vector, a normalized fifth modal feature vector and a normalized sixth modal feature vector are obtained; and feature fusion is performed on the normalized first modal feature vector, the normalized second modal feature vector, the normalized third modal feature vector, the normalized fourth modal feature vector, the normalized fifth modal feature vector and the normalized sixth modal feature vector through a deep belief network to obtain features after feature processing.

[0086] Optionally, the three-dimensional imaging reconstruction module is specifically used to: divide the internal space of the power transmission connection fitting into multiple voxel units; determine the initial acoustic parameters of each voxel unit in the multiple voxel units based on the propagation information of the acoustic signal combined with the set sound wave propagation physical model; input the initial acoustic parameters and the features after feature processing into a pre-trained deep learning image reconstruction network to obtain the target acoustic parameters of each voxel unit; map the target acoustic parameters of each voxel unit according to the three-dimensional spatial position of each voxel unit to generate a three-dimensional acoustic image.

[0087] Optionally, the defect recognition module is specifically used to: divide the three-dimensional acoustic image into two-dimensional slice images; based on a set defect recognition model, perform defect recognition on the two-dimensional slice images to obtain the defect type; wherein the set defect recognition model is a convolutional neural network model.

[0088] Optionally, the above-mentioned device also includes a data transmission module, which is specifically used to: annotate the three-dimensional acoustic image according to the annotation information to generate a defect analysis report; wherein the annotation information at least includes the defect type; transmit the three-dimensional acoustic image and the defect analysis report to a remote server, so that the remote server can analyze the three-dimensional acoustic image based on the defect analysis report.

[0089] Figure 4A schematic diagram of the structure of an electronic device 10 that can be used to implement an embodiment of the present application is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present application described and / or required herein.

[0090] like Figure 4 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., which is communicatively connected to the at least one processor 11. The memory stores a computer program that can be executed by the at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 into the random access memory (RAM) 13. Various programs and data required for the operation of the electronic device 10 can also be stored in the random access memory (RAM) 13. The processor 11, the read-only memory (ROM) 12, and the random access memory (RAM) 13 are connected to each other via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0091] Various components in the electronic device 10 are connected to an input / output (I / O) interface 15, including an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a magnetic disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0092] Processor 11 can be any general-purpose and / or specialized processing component with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various specialized artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, digital signal processors (DSPs), and any other suitable processor, controller, microcontroller, etc. Processor 11 executes the various methods and processes described above, such as the method for identifying defects in voltage-transmitting connectors.

[0093] In some embodiments, the method for identifying defects in power transmission connectors may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 10 via a read-only memory (ROM) 12 and / or a communication unit 19. When the computer program is loaded into a random access memory (RAM) 13 and executed by the processor 11, one or more steps of the method for identifying defects in power transmission connectors described above may be performed. Alternatively, in other embodiments, the processor 11 may be configured to execute the method for identifying defects in power transmission connectors in any other appropriate manner (e.g., by means of firmware).

[0094] Various embodiments of the systems and techniques described above can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), system-on-chip systems (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include being implemented in one or more computer programs that are executable and / or interpreted on a programmable system that includes at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0095] Computer programs for implementing the methods of the present application may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when the computer program is executed by the processor, the functions / operations specified in the flowcharts and / or block diagrams are implemented. The computer program may be executed entirely on the machine, partially on the machine, as a stand-alone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0096] In the context of the present application, a computer-readable storage medium can be a tangible medium that can contain or store a computer program for use by an instruction execution system, device or equipment or used in combination with an instruction execution system, device or equipment. A computer-readable storage medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium can be a machine-readable signal medium. A more specific example of a machine-readable storage medium can include an electrical connection based on one or more lines, a portable computer 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 foregoing.

[0097] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0098] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0099] A computing system may include clients and servers. The clients and servers are typically remote from each other and typically interact via a communication network. This client-server relationship arises through computer programs running on the respective computers, creating a client-server relationship. The server may be a cloud server, also known as a cloud computing server or cloud host. This server is a hosting product within the cloud computing service ecosystem that addresses the management difficulties and limited scalability of traditional physical hosting and VPS services.

[0100] An embodiment of the present application further provides a computer program product, including a computer program, which, when executed by a processor, implements the defect identification method for power transmission connection fittings as provided in any embodiment of the present application.

[0101] The computer program product, during implementation, may be written in one or more programming languages ​​or a combination thereof, for performing the operations of the present application, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0102] Note that the above are only preferred embodiments of the present application and the technical principles employed. Those skilled in the art will understand that the present application is not limited to the specific embodiments herein, and that various obvious changes, readjustments, and substitutions can be made by those skilled in the art without departing from the scope of protection of the present application. Therefore, although the present application has been described in more detail through the above embodiments, the present application is not limited to the above embodiments and may include many other equivalent embodiments without departing from the scope of the present application. The scope of the present application is determined by the scope of the appended claims.

Claims

1. A method for identifying defects in power transmission connection fittings, characterized in that: include: collecting acoustic signals of the voltage transmission connection fittings; Performing noise reduction processing on the acoustic signal to obtain a noise-reduced acoustic signal; Performing feature processing on the noise-reduced acoustic signal to obtain features after feature processing; Performing three-dimensional imaging reconstruction on the features after the feature processing to obtain a three-dimensional acoustic image; Defect identification is performed on the three-dimensional acoustic image to obtain the defect type.

2. The method according to claim 1, characterized in that Collecting the acoustic signal of the voltage transmission connection fitting includes: Transmitting continuous ultrasonic signals to the voltage transmission connection fitting according to adjustment parameters of the transmitting device; wherein the adjustment parameters include at least one of the following: operating frequency, array element excitation timing, and transmission angle; A plurality of acoustic signals are collected by an acquisition device and at a set sampling frequency; wherein the acoustic signals are reflected wave signals and transmitted wave signals corresponding to the ultrasonic signals; and the spatial position information and depth information inside the voltage transmission contact fitting corresponding to each acoustic signal are different.

3. The method according to claim 2, characterized in that in, The transmitting device is a phased array acoustic probe; the surface of the transmitting device and the surface of the voltage transmission connection fitting include a set coupling agent.

4. The method according to claim 2, characterized in that in, The acquisition device is an acoustic sensor array; the acoustic sensor array is composed of a plurality of acoustic sensors distributed around the voltage transmission connection fitting according to a set shape.

5. The method according to claim 1, wherein Performing noise reduction processing on the acoustic signal to obtain a noise-reduced acoustic signal includes: Decomposing the acoustic signal into a plurality of submodal components; wherein the plurality of submodal components have different frequency characteristics; Noise reduction is performed on each of the multiple submodal components based on local signal statistical information and filtering parameters to obtain a noise-reduced acoustic signal.

6. The method according to claim 1, wherein in, The de-noised acoustic signal includes a reflected wave signal and a transmitted wave signal; and feature processing is performed on the de-noised acoustic signal to obtain features after feature processing, including: Extracting the time domain features, frequency domain features, and time-frequency domain features of the reflected wave signal respectively, and obtaining a first modal feature vector, a second modal feature vector, and a third modal feature vector accordingly; Extracting the time domain features, frequency domain features, and time-frequency domain features of the transmitted wave signal respectively, and obtaining a fourth modal eigenvector, a fifth modal eigenvector, and a sixth modal eigenvector accordingly; Normalizing the first modal eigenvector, the second modal eigenvector, the third modal eigenvector, the fourth modal eigenvector, the fifth modal eigenvector, and the sixth modal eigenvector to correspondingly obtain a normalized first modal eigenvector, a normalized second modal eigenvector, a normalized third modal eigenvector, a normalized fourth modal eigenvector, a normalized fifth modal eigenvector, and a normalized sixth modal eigenvector; The normalized first modal feature vector, the normalized second modal feature vector, the normalized third modal feature vector, the normalized fourth modal feature vector, the normalized fifth modal feature vector and the normalized sixth modal feature vector are subjected to feature fusion through a deep belief network to obtain features after feature processing.

7. The method according to claim 1, characterized in that Performing three-dimensional imaging reconstruction on the features after feature processing to obtain a three-dimensional acoustic image includes: Dividing the internal space of the voltage transmission connection fitting into a plurality of voxel units; Determining initial acoustic parameters of each voxel unit in the plurality of voxel units based on the propagation information of the acoustic signal and a set sound wave propagation physical model; Inputting the initial acoustic parameters and the features after feature processing into a pre-trained deep learning image reconstruction network to obtain the target acoustic parameters of each voxel unit; The target acoustic parameter of each voxel unit is mapped correspondingly according to the three-dimensional spatial position of each voxel unit to generate a three-dimensional acoustic image.

8. The method according to claim 1, characterized in that Performing defect recognition on the three-dimensional acoustic image to obtain the defect type includes: dividing the three-dimensional acoustic image into two-dimensional slice images; Based on a set defect recognition model, defect recognition is performed on the two-dimensional slice image to obtain the defect type; wherein the set defect recognition model is a convolutional neural network model.

9. The method according to claim 1, characterized in that After performing defect recognition on the three-dimensional acoustic image and obtaining the defect type, the method further includes: Annotating the three-dimensional acoustic image according to the annotation information to generate a defect analysis report; wherein the annotation information at least includes the defect type; The three-dimensional acoustic image and the defect analysis report are transmitted to a remote server, so that the remote server analyzes the three-dimensional acoustic image based on the defect analysis report.

10. A device for identifying defects in power transmission connection fittings, characterized in that: include: An acoustic signal acquisition module, used to acquire the acoustic signal of the voltage transmission connection fitting; A noise reduction processing module, configured to perform noise reduction processing on the acoustic signal to obtain a noise-reduced acoustic signal; A feature processing module, configured to perform feature processing on the noise-reduced acoustic signal to obtain features after feature processing; A three-dimensional imaging reconstruction module, configured to perform three-dimensional imaging reconstruction on the features after feature processing to obtain a three-dimensional acoustic image; The defect recognition module is used to perform defect recognition on the three-dimensional acoustic image to obtain the defect type.

Images