1099 results about "Magnetic leakage" patented technology

The invention provides a rotor iron core, a permanent magnet motor and an inverter compressor, and relates to the technical field of motor and compressor design and manufacture, the rotor iron core comprises uniformly distributed permanent magnet grooves, a groove is arranged between two adjacent permanent magnet grooves, each groove extends inwards along the radial direction of the rotor iron core, parallel magnetic bridges are arranged between the grooves and the permanent magnet grooves, and the permanent magnet grooves are connected with the parallel magnetic bridges. A first magnetic barrier groove, a second magnetic barrier groove and a third magnetic barrier groove which are symmetrically distributed are formed in the two sides of the top of each permanent magnet groove respectively, a first arc and a second arc are arranged on the outer side of the rotor core above the magnetic barrier grooves, the second arc intersects with the side edge of the groove, and permanent magnets with alternating magnetism are embedded in the permanent magnet grooves. The radius Rr of the outer side of the rotor iron core, the distance ho from the middle point of the bottom of the permanent magnet groove to the center point of the rotor iron core and the thickness hm of the permanent magnet meet the condition that (Rr-ho) / hm is larger than 2 and smaller than or equal to 3.4. By reasonably arranging the position and the size of the magnetic barrier groove, magnetic circuit distribution can be changed, magnetic flux leakage can be reduced, the power density of the motor can be increased, and electromagnetic vibration noise of the motor and the compressor can be reduced.

The invention relates to the technical field of nondestructive testing, and discloses a bimetal composite pipe three-dimensional reconstruction method and system based on multi-source data fusion. The method comprises the following steps: acquiring magnetic flux leakage signals and thickness data of the bimetal composite pipe in a high-pressure environment, and preprocessing the magnetic flux leakage signals and the thickness data to obtain a clean multi-source signal data set; performing time domain and frequency domain feature alignment on the data set to generate a fusion data matrix; extracting a preliminary defect feature set through multi-layer convolution processing; performing classification training on the defect features to obtain a defect type classification result containing confidence scores; if the crack exists, depth fitting is carried out to quantify the crack depth; if the preset risk threshold value is exceeded, generating a three-dimensional defect distribution model and evaluating connectivity; and finally, outputting a quantitative evaluation report of the pipeline risk level. According to the method, efficient fusion of multi-source data, intelligent identification of defect types and three-dimensional visual reconstruction are realized, and the accuracy and evaluation efficiency of pipeline defect detection are remarkably improved.

The invention discloses an elevator steel wire rope defect detection device and method combining AR coding and magnetic flux leakage. The elevator steel wire rope defect detection device comprises a magnetic flux leakage detection sensor array device and a steel wire rope movement robot. The magnetic flux leakage detection sensor array device is installed on the steel wire rope movement robot and used for obtaining magnetic flux leakage signals of a steel wire rope in real time. And the steel wire rope movement robot is used for carrying the high-precision industrial camera and the magnetic flux leakage detection sensor array and moving along the steel wire rope to realize full-coverage detection of the steel wire rope. According to the invention, through excessive sensor data fusion, a deep learning algorithm and a virtual information generation technology, high-precision and real-time detection and positioning of elevator steel wire rope defects are realized, and the detection efficiency is remarkably improved. The device not only can detect the broken wire of the elevator steel wire rope, but also can detect the crack and stress change conditions of the steel wire rope, and gives an early warning for the safety condition of the elevator steel wire rope.

The invention relates to an anti-offset wireless charging coupler for an underwater vehicle, which solves the problem that the multi-dimensional offset charging efficiency is reduced due to ocean current by adopting the collaborative design of a grouped serial-wound transmitting end and a geometric adaptive receiving end. The transmitting end is formed by connecting an outer ring winding and an inner ring winding in series, the outer ring provides wide-area magnetic field coverage, the inner ring strengthens the local magnetic field intensity, and the ferrite array is combined to suppress magnetic leakage to cope with horizontal and rotary offset; the receiving end adopts a curved surface type structure and fits the geometric contour of an aircraft shell, interference of a rotation offset delta phi on a magnetic field is limited through optimal design of a coverage angle theta, and magnetic flux is concentrated by a ferrite array on the back side to improve the coupling efficiency. The system does not need complex mechanical adjustment, realizes energy transmission through magnetic field coupling, keeps stable efficiency in seawater, fresh water and air environments, has the advantages of compact structure, high offset resistance, wide medium adaptability and high expandability, and remarkably improves the reliability and cruising ability of wireless charging of the underwater vehicle.

The invention provides an axial magnetic field modulation type magnetic gear applying a T-shaped modulation ring structure, which belongs to the field of motors and comprises a high-speed permanent magnet rotor, a low-speed permanent magnet rotor and a T-shaped modulation ring. The modulation ring comprises a plurality of T-shaped magnetic modulation pole pieces and T-shaped non-magnetic composite material pole pieces which are alternately arranged and bonded in the circumferential direction. The high-speed permanent magnet rotor and the low-speed permanent magnet rotor each comprise a rotor iron core composed of a plurality of rotor yokes, and N-pole permanent magnets and S-pole permanent magnets embedded between the adjacent rotor yokes in the circumferential direction. The high-speed permanent magnet rotor and the low-speed permanent magnet rotor coaxially sleeve the modulation ring on the two sides of the modulation ring respectively, and form radial and axial air gaps with the modulation ring. Leakage flux at the inner peripheral ends of the rotor yokes of the high-speed and low-speed permanent magnet rotors can enter the T-shaped modulation rings to be linked with magnetic flux generated by the corresponding rotor permanent magnets, the leakage flux at the inner peripheral ends of the rotor yokes on the two sides is reduced, the high-speed and low-speed rotors coaxially rotate, the air gap flux density can be increased, and therefore the gear output torque and the torque density are improved; meanwhile, the modulation ring is easy to process and compact in structure.

The invention discloses an intelligent detection method for defects in a pipeline based on multi-sensor fusion and deep learning, and relates to the field of industrial defect detection. According to the technical scheme, three-dimensional point cloud data and two-dimensional image data in a pipeline are collected; performing synchronous positioning and map construction processing on the three-dimensional point cloud data and the two-dimensional image data in the pipeline to obtain inspection path data; performing independent feature extraction processing on the inspection path data to obtain visual, magnetic flux leakage and ultrasonic feature vector data; performing uncertainty coefficient quantification processing on the visual, magnetic flux leakage and ultrasonic feature vector data to obtain corrected joint feature data; performing graph neural network multi-level Bayesian iterative calculation processing on the corrected joint feature data to obtain global defect judgment probability data; and processing the defect judgment probability data associated with the pose data to obtain three-dimensional coordinate mapping to generate defect thermodynamic diagram data. According to the invention, the problems of single detection, insufficient precision and low path planning capability in the prior art are solved.

According to the method for pipeline magnetic flux leakage data enhancement and defect and abnormal signal detection, the deep learning network model is used to construct the data enhancement network model, the training data set is established in advance to train the data enhancement model, the model fully understands the defect characteristics in the pipeline magnetic flux leakage signal data, and the defect and abnormal signal detection accuracy is improved. Subsequently, training a ResNet101 network by using the enhanced pipeline magnetic flux leakage data set to predict the probability that the three-axis component of the target pipeline magnetic flux leakage data is a defect three-axis component, and directly carrying out defect detection on to-be-detected pipeline magnetic flux leakage signal data according to the trained ResNet101 network and a pipeline defect and anomaly judgment criterion. The method has the advantages that the method is simple, manual experience and knowledge storage are not needed, data features are automatically extracted, defects of pipeline magnetic flux leakage signal data are automatically detected, and the problems that a deep model needs a large amount of data to be trained, the actually-collected historical pipeline magnetic flux leakage data amount is seriously insufficient, and sufficient training of a deep learning model cannot be supported are solved.

The invention discloses a pipeline defect magnetic flux leakage detection method and device based on multi-scale data driving deep learning, and relates to the technical field of pipeline defect detection. A multi-scale magnetic flux leakage signal data set containing defect global distribution and local details is generated, and multi-level primary features are automatically extracted by using a convolutional neural network; and the defects of poor generalization and easy key information omission of a manual method are overcome. And then dynamically enhancing and performing weighted fusion on multi-scale features by means of a multi-scale convolution branch and an attention mechanism, so that the model can learn global and local features of the defect at the same time, the problem that the global and local features of the defect are difficult to consider in traditional deep learning is solved, and through transverse connection and up-sampling fusion of a feature pyramid, a multi-scale feature is obtained. And the features have high semantic information and high spatial details. And finally, independently carrying out multi-task prediction by virtue of a task decoupling module, and calibrating result consistency by virtue of a task alignment module, so that the detection accuracy in a complex scene is improved, and more accurate and robust pipeline defect magnetic flux leakage detection is realized.

The invention discloses a magnetoacoustic coupling detection system and method for pipeline defect detection, and relates to the technical field of pipeline nondestructive detection. The system comprises a pipeline defect detection module, a magnetofluid driving module, an auxiliary moving module, a signal acquisition module, a data processing module and a data transmission module. The pipeline defect detection module comprises a magnetic flux leakage detection unit and an ultrasonic detection unit; the magnetofluid driving module comprises magnetofluid and an electromagnetic excitation device; the auxiliary moving module comprises a plurality of driving wheels tightly attached to the inner wall of the pipeline. The signal acquisition module comprises a data acquisition unit and a data storage unit; and the data processing module is used for processing the acquired magnetic excitation parameters, the magnetic leakage signals and the sound wave sensing data. According to the invention, the technical problems of low defect identification precision, complex use of a coupling agent, insufficient real-time data processing capability and the like existing in a single detection mode in the prior art can be solved.

Disclosed in the present invention are a sound production device, a sound production module, and an electronic apparatus. The sound production device comprises a magnetic circuit system, an electrically conductive support and a vibration system, wherein the magnetic circuit system comprises a magnetically conductive yoke, and a central magnetic part and an edge magnetic part arranged on the magnetically conductive yoke, the edge magnetic part and the central magnetic part enclosing a magnetic gap; the magnetic circuit system is provided with a mounting hole running through the magnetically conductive yoke and the central magnetic part, the electrically conductive support being arranged in the mounting hole, and the electrically conductive support being provided with a first welding surface; and the vibration system comprises a diaphragm assembly, a voice coil and a spider which are opposite the magnetic circuit system, one end of the voice coil being connected to the diaphragm assembly, and the other end of the voice coil being suspended in the magnetic gap, one end of the spider being electrically connected to the first welding surface, and the other end of the spider being connected to the diaphragm assembly and electrically connected to a lead-out wire of the voice coil. The first welding surface is not higher than the side surface of the central magnetic part facing the diaphragm assembly. The sound production device of the present invention has low magnetic leakage, so that the magnetic field intensity can be effectively improved, thus increasing the BL value, improving the acoustic performance.

The invention discloses an oil and gas pipeline magnetic flux leakage image defect identification method based on a deep attention mechanism, relates to the technical field of oil and gas pipeline detection, and is used for solving the problem of inaccurate identification of a magnetic flux leakage image of a pipeline elbow section. According to the method, the image frame sequence with the posture annotation is constructed through unified time reference and space coordinate mapping, and accurate alignment of the image and the pipeline position is achieved; a structural area marking graph and non-rigid normalization are introduced to compensate the distortion of the elbow section, and the image consistency is improved; constructing a structure perception embedded image on the distortion compensation image, fusing position, gradient and texture features to implement feature propagation and attention guidance, and generating a feature saliency map; a defect area is accurately extracted in combination with layered reconstruction and a two-stage judgment strategy; through continuous frame monitoring and recognition stability judgment, recognition parameters are adaptively updated, a closed loop from recognition to updating to verification is constructed, and the precision of oil and gas pipeline magnetic flux leakage image recognition under complex working conditions is enhanced.

The invention relates to the field of state monitoring and fault diagnosis of power equipment, in particular to a transformer fault diagnosis method based on a chaos evolutionary optimization algorithm, which comprises the following steps of: 1, acquiring a magnetic flux leakage signal during operation of a transformer; 2, optimizing a parameter modal number K and a penalty factor alpha of variational modal decomposition by using a chaos evolutionary optimization algorithm; 3, performing variational mode decomposition on the magnetic flux leakage signal to obtain an intrinsic mode function component; 4, calculating the envelope entropy of the intrinsic mode function component, and obtaining an effective intrinsic mode function component through screening; 5, extracting the energy entropy and the sample entropy of the effective intrinsic mode function component to form a feature vector; and 6, inputting the feature vector into a pre-trained support vector machine classifier, and outputting a fault type diagnosis result of the transformer. According to the method, the CEO algorithm is combined with the ergodicity of chaotic mapping and the global search capability of the evolutionary algorithm, and the problems that VMD parameters K and alpha are sensitive and depend on experience, and a traditional optimization algorithm is prone to local optimum are effectively solved.

The invention relates to the technical field of pipeline defect detection, and discloses a magnetic flux leakage data defect identification and detection method and system. The method comprises the following steps: acquiring a multi-channel magnetic flux leakage detection signal of a to-be-detected pipeline and spatial position information of each detection node; determining time sequence offset according to spatial position correlation characteristics of adjacent detection nodes and magnetic flux leakage signal propagation characteristics, performing space-time alignment on the magnetic flux leakage signals, and extracting signal distribution characteristics in a preset analysis window; then combining the signal distribution characteristic difference degree of the adjacent detection nodes and the defect characteristic fluctuation factor to construct a defect characteristic incidence matrix, analyzing the matrix to obtain defect probability distribution of different frequency bands of each detection node, and extracting a target defect characteristic frequency; secondly, calculating a defect characteristic nonlinear error probability according to the spatial distribution consistency of target defect characteristic frequency, and dynamically adjusting a classification decision threshold parameter of a defect recognition model according to the defect characteristic nonlinear error probability; and finally, completing automatic identification and classification of pipeline defects by using the adjusted model.

The invention discloses an MFL signal reconstruction method based on channel mutual fusion and a global attention mechanism, relates to the technical field of nondestructive testing, and particularly discloses a magnetic flux leakage signal MFL reconstruction method based on a neural network. In order to solve the problem of interference of external noise on defect characteristic signals, a signal channel recombination, shuffling and aggregation optimization method based on CycleGAN is adopted, a GSoP optimization covariance matrix is introduced, through PSNR and LPIPS index comparison, the reconstruction precision of the MFL defect signals can be greatly improved, and the reconstruction precision of the MFL defect signals is improved. And the data processing precision and speed are improved (the time cost and the space cost of data processing are reduced), so that the defect detection efficiency is improved.

The utility model discloses a rotor, motor and compressor relates to motor equipment technical field, rotor includes permanent magnet and rotor iron core, rotor iron core has rotation axis, rotor iron core includes a plurality of laminations, the plurality of laminations are laminated along rotation axis, each lamination forms a plurality of magnetic slot, and the plurality of magnetic slot forms the permanent magnet. The magnetic slots of the plurality of laminations are communicated with each other along the direction of the rotating axis so as to form an accommodating slot for accommodating the permanent magnet; each lamination forms a magnetic bridge or a communication groove at one magnetic groove, the magnetic bridge is located at the edge of the lamination and blocks the magnetic groove, and the communication groove is communicated with the magnetic groove and the external space; the number of the laminations is defined as N, the number of the magnetic grooves in every two adjacent containing grooves is 2N, the number of the magnetic bridges is M, and the ratio of M to 2N ranges from 0.2 to 0.5. The utility model aims to reduce the magnetic flux leakage of the rotor iron core on the basis of ensuring the structural strength through the rotor iron core so as to improve the motor efficiency.

The invention relates to the technical field of magnetic suspension bearings, and provides a magnetic leakage prevention bearing structure which comprises a flux sleeve, a shell, a magnetic part, an inner-layer magnet yoke and an outer-layer magnet yoke. The shell is sleeved outside the flux sleeve, and an inner ring of the shell is provided with a mounting space; the magnetic part is arranged in the mounting space; the inner-layer magnet yoke is arranged in the mounting space and connected with the shell, the inner-layer magnet yoke is concaved inwards to form a first containing groove, and the magnetic part is contained in the first containing groove; the outer-layer magnet yoke is arranged in the installation space and connected with the shell, the outer-layer magnet yoke is concaved inwards to form a second containing groove, and the inner-layer magnet yoke is contained in the second containing groove. A closed magnetic circuit is formed through a double-layer magnet yoke nested structure, the axial magnetic shielding layer design is matched, magnetic flux leakage is effectively restrained, the magnetic field energy utilization rate is increased, electromagnetic interference to external equipment is reduced, and meanwhile the bearing performance and operation stability of the bearing are remarkably improved by optimizing magnet yoke thickness distribution and the permanent magnet arrangement mode.

The utility model discloses a detection device suitable for oil gas pipeline magnetic flux leakage, which comprises a probe magnet section, three groups of universal joints, a battery section and a pulley, the universal joints are arranged on two sides of the probe magnet section, the battery section is arranged on the right side of the universal joints, another group of universal joints is arranged on the right side of the battery section, and the pulley is arranged on the other group of universal joints. Three groups of pulleys are arranged on the right side of the universal joint, and a cleaning assembly is arranged at the left end of the universal joint on the left side of the probe magnet section. A cleaning assembly is arranged at one end of a universal joint on the left side of a probe magnet joint, so that an extrusion plate in the cleaning assembly rotates on a threaded rod in a threaded mode and extrudes a rotating rod, the rotating rod rotates in a fixing block and drives a scraper to be attached to the inner wall of a pipeline, and then a transmission rod is driven by a motor to rotate; and the scraper blade is used for cleaning attachments on the inner wall of the pipeline, so that interference on the probe magnet section during use is reduced, and the accuracy of a detection result of the probe magnet section is further improved.

The present invention relates to a new type of solenoid valve in the field of solenoid valve technology, comprising a valve body and a valve mechanism, wherein an inlet channel and an outlet channel are arranged in the valve body, the valve mechanism comprises a seal for isolating the valve cavity and an electromagnetic structure for driving the seal to move, the electromagnetic structure comprises a cover body, a magnet and a coil, and the magnet and the coil are installed in the cover body at intervals. The present invention embeds the magnet and the coil in the cover body, thereby greatly reducing magnetic leakage, improving the magnetic field utilization rate of the coil, reducing the number of coil turns, reducing power consumption, and effectively solving the heating problem of the solenoid valve. At the same time, the cover body can effectively isolate the external magnetic field, avoid interference with the internal electromagnetic structure, and ensure the stability of the working operation of the solenoid valve; the coil and the valve core with cantilever support are arranged in the gap between the isolation cover and the magnet, so that the solenoid valve will not have any mechanical friction during the process of driving the switch, and the problems of wear and mechanical noise of the solenoid valve are completely solved.

The invention discloses a steel wire rope nondestructive testing method and system based on sensing and physical feature fusion, and belongs to the technical field of steel wire rope nondestructive testing. The method comprises the following steps: firstly, carrying out saturation magnetization on a steel wire rope through an axial Gaussian difference excitation probe so as to improve the signal-to-noise ratio of a defect magnetic flux leakage signal; and meanwhile, the instantaneous acceleration is synchronously acquired through the inertial measurement unit. And establishing a speed-signal model based on the law of electromagnetic induction, and performing dynamic speed compensation on the magnetic flux leakage original voltage signal. And then, quantitatively extracting characteristic parameters with clear physical significance, such as peak voltage, maximum gradient, full width at half maximum and the like, from the standardized magnetic flux leakage signal, fusing the characteristic parameters with deep characteristics extracted by the deep learning model, and inputting a fused characteristic vector into a full-connection neural network and a classifier to realize defect type identification and degree evaluation. According to the invention, high-precision and interpretable intelligent detection and residual life prediction of steel wire rope defects can be realized, the detection reliability is obviously improved, and predictive maintenance is supported.

The invention discloses a magnetic flux leakage internal detection system for gas pipe network detection and a control method, and mainly relates to the technical field of gas pipeline internal detection. Comprising a main power section, a magnetic leakage section, an auxiliary power section and a universal joint, the main power section comprises a power section sealing cabin, the power section sealing cabin is of a hollow cylindrical structure, an odometer wheel module is arranged on the outer side wall of the main power section sealing cabin, and a control circuit support and a battery control support are arranged in the main power section; the two ends of the main power section are provided with an end cover for debugging and a rear end cover respectively, the rear end cover and the magnetic leakage section are connected through a universal joint, and the magnetic leakage section and the auxiliary power section are connected through a universal joint. The device has the beneficial effects that the walking friction resistance of the magnetic flux leakage inner detector is reduced, and the bidirectional movement in a pipe network is also realized.

The invention relates to the technical field of image processing, and provides a pipeline magnetic flux leakage signal metal loss detection method based on weak labeling, and the method comprises the steps: converting an obtained magnetic flux leakage signal image into an RGB image, training a detection model through employing a preset number of RGB images with rectangular frame labeling, enabling the detection model to generate a prediction rectangular frame, and carrying out the prediction of the prediction rectangular frame; training a classification model by using the RGB image with the category label so as to enable the classification model to obtain a category confidence score; generating a thermodynamic diagram based on the classification model, generating a fitting rectangular frame based on the thermodynamic diagram, and calculating a fitting score in combination with an overlapping degree evaluation result of the prediction rectangular frame and the fitting rectangular frame and the category confidence score; and obtaining a target detection model through the fitting score to execute metal loss detection on the to-be-detected magnetic signal image to obtain a detection result. According to the method, the detection model and the classification model are cooperated, and the thermodynamic diagram and the fitting evaluation are combined, so that the metal loss detection precision is improved while the manual labeling requirement is reduced.

The invention relates to the technical field of electric signal data processing, in particular to a cable life prediction method based on working condition movement. The method comprises the following steps of: sleeving a leakage flux ring on the periphery of a cable shielding layer, wherein a Hall array is arranged in the leakage flux ring; an acoustic emission patch is attached to the surface of the shielding layer; a mechanical trigger is arranged at the joint of the cable drag chain; when the mechanical trigger detects a complete bending stroke every time, the Hall array and the acoustic emission patch are started synchronously, and a magnetic flux leakage pulse amplitude value and an ultrasonic elastic wave energy value of the current stroke number are generated; dividing the magnetic flux leakage pulse amplitude value of the same stroke number by the ultrasonic elastic wave energy value to obtain a broken wire coefficient of the stroke number; and accumulating the broken wire coefficients of the continuous stroke numbers of the cable to form a theoretical accumulated broken wire index. According to the invention, through the magnetic flux leakage and acoustic emission monitoring technology and in combination with the accelerated bending test, accurate quantification and life prediction of the broken wire state of the cable are realized, and finally, the reliability of the whole life cycle management of the cable is improved.

The invention discloses a novel permanent magnetic mechanism suitable for a permanent magnetic circuit breaker and a ring main unit, which is characterized in that a static iron core and a movable iron core are oppositely arranged in an aluminum alloy shell along the axial direction, an internal pull rod penetrates through the axis of the aluminum alloy shell, and the internal pull rod movably penetrates through the static iron core and is connected with the movable iron core; a shaft fixing ring is movably arranged on an inner pull rod of the aluminum alloy shell, the shaft fixing ring movably abuts against the side, away from the static iron core, of the movable iron core, and the two sides of the shaft fixing ring are axially limited through a locking nut and a total stroke adjusting nut which are connected to the inner pull rod in a threaded mode. The mechanism is provided with a total stroke adjusting mechanism, an overtravel adjusting mechanism and a switching-on and switching-off speed adjusting mechanism, respective adjustment is achieved through special nuts, the problem of synchronous change of the opening distance and the overtravel of a traditional mechanism is solved, an aluminum alloy shell and a closed magnetic circuit are adopted, magnetic leakage is reduced, the magnetic field utilization rate and the impact resistance are improved, and spring pre-tightening force is changed through spring adjusting nuts; the switch-on and switch-off speed is accurately controlled, various loads are adapted, and the breaking risk is reduced.

The invention discloses a coil structure-based length self-sensing method for high polymer twisted artificial muscle, which comprises the following steps of: firstly, enabling the artificial muscle to be equivalent to N equidistant circular current loops, establishing an inductance calculation formula by combining a Biot-Savart law and a magnetic field superposition principle, and breaking through the limitation of a traditional solenoid model on a helical angle; then through MATLAB numerical simulation and iterative computation, fitting to obtain a power function describing the relation between inductance and length; and finally, proposing a parameter dynamic correction mode based on single-time initial inductance measurement according to complex factors such as actual magnetic flux leakage and skin effect, and realizing rapid adaptive calibration by adjusting a parameter b. In a word, the method creatively combines theoretical modeling with experimental correction, the sensing precision in a dynamic stretching scene is remarkably improved while the inductance-length macroscopic trend prediction capability is ensured, dependence on a large amount of experimental data is not needed, and the method has good engineering applicability and is worthy of popularization.

A magnetic adsorption wheel type cleaning, double-excitation magnetic flux leakage and ultrasonic thickness measurement detection robot comprises a vehicle body and an upper cover. A cleaning device for cleaning the wall surface is arranged at the front end of the vehicle body; the upper cover is provided with a shear type lifting mechanism used for adjusting the distance between the magnetizing device and the detected object, and the shear type lifting mechanism changes the distance between the magnetizing device and the detected object, so that the magnetic field intensity is adjusted; the magnetizing device is carried in the magnetic adsorption wheeled robot body and is responsible for magnetizing a detected object; the rear end of the vehicle body is provided with a thickness measuring module for detecting the thickness of a detected object. The vehicle body chassis is provided with a magnetic wheel adsorption device, and the magnetic wheel adsorption device drives the magnetic adsorption wheel type magnetic flux leakage detection robot; a probe fine adjustment mechanism is arranged on the side face of the vehicle body, the probe fine adjustment mechanism achieves accurate movement of the probe in the vertical direction, and therefore the detection probe and the detected surface are located at the optimal detection distance. According to the invention, the sensitivity, accuracy and adaptability of magnetic flux leakage detection can be improved, so that wall surface defects can be detected.

The invention discloses an electromagnetic transducer modeling and dynamic distribution parameter confirmation method considering electro-magnetic-mechanical-acoustic coupling. The method comprises the following steps: constructing a dynamic multi-physical field distribution parameter model DPM of an electromagnetic transducer system; a PGA model optimization method based on a parallel genetic algorithm; the dynamic multi-physical field distribution parameter model DPM comprises the following steps: constructing an excitation loop equivalent model, wherein the excitation loop equivalent model comprises a plurality of excitation main loop units, a plurality of air gap magnetic flux leakage loop units and a plurality of permanent magnet loop units; and constructing a mechanical structure equivalent model, wherein the mechanical structure equivalent model comprises physical models such as a system radiation panel, a spring system and a rubber panel. According to the method, parameters of the DPM are optimized through the PGA algorithm, the performance efficiency of the electromagnetic transducer is improved, and a new method is provided for design of the electromagnetic transducer.

The invention discloses a defect detection system based on a bias alternating current magnetization modulation TMR sensor, and the system is characterized in that through the cooperation of an alternating current and direct current magnetization module, a TMR sensor and a defect prediction module, the alternating current and direct current magnetization module is used for simulating an alternating current and direct current superimposed magnetic field test environment, and a reverse alternating current and direct current magnetic saturation magnetic field is formed; the TMR sensor obtains defect signal characteristics of a to-be-detected workpiece and sends the obtained defect signal characteristics of the to-be-detected workpiece to the defect prediction module, and the defect prediction module achieves full-scale detection of defects from 20 micrometers to millimeter by analyzing peak values (micron-scale narrow defects) and waveform widths (millimeter-scale defects) of alternating-current waveforms. The balance of high sensitivity and wide range of the same sensor is realized, and meanwhile, the dual requirements on precision and efficiency in actual detection are met. Meanwhile, through alternating current and direct current magnetic field cooperative modulation, the measuring range TMR sensor is expanded to 300 Gs, and the full range from weak magnetic leakage to strong magnetic leakage is covered.

The invention discloses a metal structure defect multi-source heterogeneous nondestructive testing signal composite characterization method based on a deep neural network and an attention mechanism. According to the method, a deep neural network model is constructed, and data fusion and defect composite characterization are carried out on multi-source heterogeneous signals obtained by three nondestructive testing technologies of magnetic flux leakage testing, electromagnetic ultrasonic testing and electromagnetic guided wave testing in combination with an attention mechanism. Firstly, through advanced signal processing and data fusion technologies, the data processing and analysis process is simplified, effective processing and interpretation of complex data are achieved, and the detection efficiency and reliability are improved; and then more accurate and comprehensive characterization of the size, type, shape and position of the metal structure defect is realized by comprehensively utilizing information of different signal sources, and innovative technical support is provided for nondestructive testing in the industrial field.

The invention relates to the technical field of signal processing, and provides a magnetic flux leakage signal noise reduction model establishment and noise reduction method based on dynamic feature fusion. The method comprises the following steps: acquiring a magnetic flux leakage signal of an irregular defect; adding noise matched with the current environment into the magnetic flux leakage signal; extracting a time domain feature and a frequency domain feature from the magnetic flux leakage signal with noise to obtain a time frequency feature; extracting features of different scales from the time-frequency features based on a self-adaptive receptive field, and performing nonlinear mapping and weighted splicing processing on the features of different scales according to self-adaptive weights to obtain spliced features; decoding and reconstructing according to the splicing features to obtain a noise reduction signal; evaluating the difference between the noise reduction signal and the magnetic flux leakage signal according to the signal-to-noise ratio index and the loss function; and updating network parameters including the dynamic weight, the adaptive weight and the receptive field by using a gradient iterative algorithm until the difference meets a preset condition, so as to establish a magnetic flux leakage signal noise reduction model. According to the embodiment of the invention, the noise reduction precision of the magnetic flux leakage signal can be improved.

The invention relates to an oil and gas pipeline magnetic flux leakage detection defect quantification method based on a composite convolutional neural network. The method comprises the following steps: 1) a data acquisition stage: carrying out saturation magnetization detection on a pipeline by adopting a pipeline internal detector to obtain magnetic flux leakage internal detection data containing defects; 2) a data preprocessing stage: performing normalization processing on the magnetic flux leakage data, and constructing a defect quantification data set; 3) a model construction stage: constructing a composite convolutional neural network model; 4) a model training stage: inputting a defect quantification data set sample into the model for training; and 5) a model test stage: performing feedback and iterative optimization for multiple times through a back propagation optimization algorithm of the model, finally outputting optimal model parameters in the training process, and exporting and storing the model parameters. Based on the pipeline magnetic flux leakage detection data, high-precision quantification of pipeline defect size parameters can be realized, and the accuracy and reliability of defect evaluation can be improved.