139 results about "Equipment Defects" patented technology

The invention belongs to the technical field of railway inspection, and particularly relates to a space-air-ground integrated railway power transformation and distribution intelligent inspection method and system which comprises a sensing layer, a platform layer, a data and intelligent center, an application layer and a user and presentation layer. The sensing layer is composed of an aerial inspection layer, a ground inspection layer and a fixed monitoring layer; the platform layer is based on a cloud computing center and edge computing node collaborative architecture, combines data and an intelligent center, integrates multi-source heterogeneous data through a big data platform, carries an AI algorithm engine to realize automatic identification of equipment defects, analyzes images, synchronously predicts the images, and realizes automatic identification of the equipment defects. And the application layer is combined to complete the functions of automatic generation of an inspection report, automatic distribution of a maintenance work order, equipment health state evaluation and the like. According to the method, functions of railway power place inspection, power supply line inspection, power supply and distribution facility operation and maintenance emergency and the like can be realized, deep fusion analysis of multi-dimensional data is realized, and meanwhile, the problems of insufficient training samples, sparse defect types and the like of an existing AI model are solved.

The invention relates to a construction method of a power grid equipment defect training sample set and a defect detection method thereof in the field of computer vision and power inspection. The construction method comprises the following steps: potential space mapping; injecting noise and conditions; performing condition denoising; defects are directionally generated. The defect-free background and structure information is reserved under the guidance of a diffusion model, and defect features consistent with text vector description are generated in a specified area, so that a piece of defect-free image is converted into an image with a known fault state for forming a power grid equipment defect training sample set. According to the method, the controllably generated diffusion model is introduced to enhance scarce defect category samples, the sample diversity is improved in combination with a traditional and adversarial generation method, and the problems of incomplete data, scarce defect samples, unbalanced category distribution, insufficient fine granularity detection precision and the like generally existing in existing power grid equipment defect detection are solved.

The invention discloses a power transformation equipment defect processing method and system based on edge-cloud collaboration, and belongs to the technical field of power system automation, the power transformation equipment defect processing method based on edge-cloud collaboration comprises the following steps: synchronously acquiring and preprocessing a visible light image and an infrared image of power transformation equipment; the edge processing layer performs equipment part positioning on the preprocessed image, extracts structural features in the visible light image and temperature features in the infrared image according to a positioning result, preliminarily screens out a suspicious area meeting a suspected defect condition, and uploads the result to the cloud analysis layer; after the cloud analysis layer receives the data, visible light features and infrared features of the visible light image and the infrared image are extracted respectively, and cross-modal feature fusion is carried out by adopting an attention mechanism. Through the edge-cloud collaborative cross-modal recognition and dynamic optimization mechanism, the problems that efficiency and precision are difficult to consider at the same time, the single-modal omission ratio is high, and the model adaptability is poor in the prior art are effectively solved.

The invention provides a hydropower station full-scene intelligent inspection and safety supervision method and device. A two-legged inspection robot with anti-explosion and anti-vibration capabilities is deployed to realize all-terrain autonomous inspection of equipment nodes in the hydropower station; based on a digital twinborn model and an edge computing platform, fusing equipment state data and personnel behavior data acquired by a robot in real time, identifying equipment defects and personnel violation behaviors through an AI algorithm, and generating graded early warning information; the robot is controlled to carry out real-time intervention through a voice module, early warning information is pushed to a safety management system, and an equipment operation authority locking mechanism is linked; through optimizing a path planning algorithm and a multi-robot cooperative scheduling strategy, rapid response and autonomous obstacle avoidance of the robot under an emergent task are realized. According to the method, full-scene automatic inspection of the hydropower station and real-time supervision of personnel behaviors can be realized, the equipment defect identification accuracy and the emergency response efficiency are remarkably improved, and the manual inspection cost and the non-planned shutdown risk are reduced.

The invention relates to the technical field of thermal power plant safety production, and discloses a thermal power plant inspection and maintenance work site risk assessment method and system based on an AI visual technology, and the method comprises the steps: obtaining the AI visual data and environment perception data of an inspection and maintenance site through a data collection module; the risk factor extraction module identifies three types of risk factors including personnel violation, equipment defects and environment abnormity from the visual data; the risk assessment module calculates the weight of each risk factor through a subjective-objective fusion weight method, obtains a dimension risk value in combination with a correction coefficient, and further solves a comprehensive risk value; and the risk early warning and disposal module divides early warning levels according to the comprehensive risk values and generates targeted disposal measures. According to the invention, automatic and precise assessment of the multi-dimensional risk of the thermal power plant inspection and maintenance site is realized, the problems of missed judgment, misjudgment and early warning lag of traditional manual inspection are solved, the safety management cost is reduced, and the safety guarantee capability of inspection and maintenance operation is improved.

The invention discloses a defect detection method and device for substation equipment, equipment and a storage medium, and relates to the technical field of substations, and the method comprises the steps: obtaining a to-be-detected image of the substation equipment, and carrying out the illumination normalization processing of the to-be-detected image, and obtaining a to-be-detected image after space alignment; generating an abnormal probability mask based on the to-be-detected image after space alignment; according to the to-be-detected image of the substation equipment, acquiring an equipment area mask through an equipment segmentation model, and performing defect target detection in an area limited by the equipment area mask to obtain a defect position, category information and a first confidence coefficient; and fusing the abnormal probability mask, the defect position, the category information and the first confidence coefficient to obtain a final defect detection report. According to the method, the defect detection accuracy of the substation equipment can be improved.

The invention discloses an equipment defect positioning method and device, a storage medium and computer equipment, and the method comprises the steps: obtaining a visible light image, an infrared image and an acoustic image of to-be-detected equipment collected by inspection equipment, extracting equipment image features from the visible light image, extracting infrared partial discharge features from the infrared image, and obtaining an infrared partial discharge feature of the to-be-detected equipment; and extracting acoustic imaging features from the acoustic imaging; performing feature fusion on the equipment image features, the infrared partial discharge features and the acoustic imaging features to obtain fusion features, and inputting the fusion features into a defect identification model to determine whether the equipment to be detected has defects; if the to-be-detected equipment has the defect, acquiring a first coordinate when the inspection equipment collects an image, and determining a second coordinate corresponding to the defect according to the first coordinate; and according to the second coordinate, positioning the defect on a three-dimensional model corresponding to the to-be-detected equipment.

The invention provides an electrical equipment defect identification method and system based on machine vision, and relates to the technical field of equipment defect identification, and the method comprises the steps: collecting and preprocessing a standard electroluminescent image, a standard infrared thermal image and a standard visible light image of a target photovoltaic string under different time windows through an imaging device carried by an unmanned aerial vehicle, and carrying out the collection of the standard electroluminescent image, the standard infrared thermal image and the standard visible light image; meanwhile, recording environmental parameters collected each time; analyzing visual defect features and attenuation gradient features in the standard electroluminescent image, and performing fusion verification by combining the standard infrared thermal image, the standard visible light image and the electrical performance data to obtain definite judgment of potential-induced attenuation; for the photovoltaic module which is diagnosed and judged to have potential-induced degradation, based on the environmental parameters, evaluating by using a time sequence analysis method to obtain the evolution rate of the potential-induced degradation, and mapping to generate the risk level of the potential-induced degradation; and generating a structured defect identification report. And the accuracy of defect identification of the power equipment is improved.

The invention discloses a GIS equipment defect intelligent detection method based on computer vision, and aims to solve the problems that transient defect acquisition in a GIS cabinet is delayed and asynchronous alignment of audio and video is difficult. According to the method, time reference and annular pre-buffering are unified, collection of a streaming voiceprint trigger linkage event camera and a high-speed polarization camera, monotonic alignment Transform of arrival time difference prior constraint and residual time offset calibration, secondary collection closed loop of confidence gating and multi-modal fusion judgment are carried out, and multi-modal fusion judgment is carried out. The technical effects of low-delay triggering, order-preserving alignment, accurate time positioning, evidence chain integrity rate improvement and missing detection and false detection reduction are realized.

The invention discloses a power transmission equipment insulator defect identification method and device, equipment and a medium, and belongs to the field of power equipment defect identification, and the method comprises the steps: respectively extracting corresponding feature data from electric signal monitoring data, infrared data and image data of an insulator string; fusing the characteristic data corresponding to the electric signal monitoring data and the infrared data through a cross-modal attention mechanism, and positioning an abnormal region; determining a defect physical region image of the insulator string according to the abnormal region and feature data corresponding to the image data; performing secondary verification on the defect physical region image according to the environmental parameter data of the space where the insulator string is located; and when the secondary verification is passed, respectively calculating a plurality of defect judgment indexes of the insulator chain according to the defect physical region image, and determining that the insulator chain has defects when any defect judgment index meets a preset condition. According to the invention, the problem of low insulator defect identification accuracy in a complex environment can be solved.

The invention relates to the technical field of equipment defect detection, and discloses an equipment defect comprehensive detection system and method based on three modes. The system comprises a multi-modal data acquisition module which acquires infrared temperature data, X-ray image data and visual image data of equipment in real time; the comprehensive data processing module processes the data and generates standardized defect features; the defect risk assessment module assesses the comprehensive risk level of the defect based on the characteristics; the detection stage division module divides the detection process into a plurality of stages and determines a detection focus of each stage; the parameter rejection degree evaluation module evaluates the rejection degree of the detection parameters based on the equipment material characteristic data and the detection focus; the adjustment amount determination module determines the adjustment amount of the detection parameter based on the rejection degree and the risk level; and the detection regulation and control module regulates and controls the detection parameters according to the adjustment amount and executes defect processing actions. According to the system, integrated detection of multiple types of defects can be realized, and the detection flexibility and practicability are improved.

The invention provides edge equipment, center equipment, a defect detection method, equipment and a medium. According to one example of the application, the edge device may include an edge detection unit and an edge management unit; wherein the edge detection unit is connected with the edge management unit and is used for acquiring detection data of a to-be-detected product based on configuration data of the to-be-detected product and generating a detection result of the detection data by utilizing a defect detection model; and the edge management unit is used for sending the detection result to the central equipment, so that the central equipment judges that the to-be-detected product has the defect problem.

The present application belongs to the technical field of glass product production management and data processing, and relates to a glass product production management method and system. The method comprises: collecting glass surface ripple signals and process parameters of a production line in real time; calculating a local curvature sequence and a local curvature entropy reflecting the sharpness of the micro-waveform of the signals, and adaptively determining a dynamic bandwidth adjustment factor of a variational mode decomposition algorithm based on the local curvature entropy; using the dynamic bandwidth adjustment factor to perform mode decomposition on the glass surface ripple signals, and extracting a characteristic component matching the frequency of a roller fault; calculating a physical real wear index after eliminating the influence of the speed according to the characteristic component and the glass drawing speed, and making a device maintenance decision based on the index. The present application can accurately identify device defects under complex working conditions by adaptively adjusting algorithm parameters and decoupling the influence of speed, thereby improving the accuracy of maintenance decisions.

The invention relates to the technical field of safety protection, in particular to a valve cooling external cold water cooling tower abnormal sound online monitoring device and safety monitoring method based on a sound sensor, and the device comprises a microphone collection module which is arranged on a target valve cooling external cold water cooling tower and is used for obtaining acoustic signals generated in the operation process of the cooling tower in real time, the acoustic signal is converted into digital audio data; the MCU mainboard module is connected with the microphone acquisition module and is used for receiving the audio data of the microphone acquisition module and realizing feature extraction, model reasoning and logic control; the communication module is used for transmitting processing information of the MCU mainboard module to a mobile terminal to realize online early warning of abnormal sound of the valve cooling external cold water cooling tower, and the device realizes real-time sound monitoring of the cooling tower, judges equipment of which the sound exceeds a normal range, and sends an alarm signal at the terminal to ensure that equipment defects are found as early as possible and treated in advance; defect deterioration is avoided.

The invention provides an infrared defect discrimination method based on an improved neural network and a related device, and the method comprises the steps: obtaining an infrared image of power inspection, and carrying out the processing through employing a constructed infrared equipment image recognition model and a temperature difference recognition model, and obtaining a target detection and temperature difference result. The temperature difference result reflects the difference between the target current temperature and the reference temperature. And based on the temperature difference judgment basis of the equipment defect, carrying out logic analysis on the detection result to determine the equipment defect type. An infrared equipment image recognition model is based on an improved YOLOv5 network, and the target recognition effect is improved by introducing an Octave Conv module and a CARAFE operator to optimize feature extraction and up-sampling stages. The improved recognition model is used for automatically processing the infrared image, manual interpretation is reduced, misjudgment caused by visual fatigue is avoided, the method is not limited by professional ability of personnel, the judgment efficiency is improved, the processing time is shortened, and the infrared heating defect problem can be solved in time.

The invention discloses an equipment defect identification method based on multi-modal information fusion and self-adaptive information optimization, and aims to solve the problems that a traditional single-modal defect identification method is difficult to effectively fuse multi-source heterogeneous information in a substation inspection log, and the model efficiency is low due to information redundancy. The method comprises the following steps: firstly, constructing a gating level attention multi-modal feature fusion network, and realizing deep integration and cross-type interaction modeling of multi-modal information; secondly, designing an information attribute importance analysis mechanism, and quantifying the contribution degree and redundancy of each information attribute to defect identification; then, constructing a self-adaptive information attribute selection framework, and realizing accurate identification through joint optimization of information attribute screening and model training; and finally, an intelligent identification system covering data processing, feature fusion, attribute optimization and defect prediction is established.

This invention relates to the field of continuous annealing furnace equipment defect detection. A method for detecting defects in continuous annealing furnace equipment based on furnace roller imprints and furnace atmosphere is provided. The method involves preparing a detection carrier by selecting a carbon steel adjusting material as the detection carrier, the width of which covers the effective working area of ​​the furnace roller (i.e., the width of the detection carrier is the maximum strip width that the continuous annealing furnace equipment can process). The furnace roller is then refurbished by passing the adjusting material through the continuous annealing furnace equipment once, controlling the refurbishment speed (running speed) to a first set speed. This forms a uniform and clear furnace roller refurbishment imprint on the surface of the adjusting material. Using this refurbishment imprint, a reference mark corresponding to the furnace roller position is established on the adjusting material, completing the preparation of the detection carrier. For continuous annealing furnace equipment defect detection, the detection carrier passes through the continuous annealing furnace equipment at a second set speed, and different types of imprints are marked with different colors.

The invention discloses a method and device for calculating the defect rate of series compensation secondary equipment under different operation years, a medium and equipment. The method comprises the following steps: extracting a series compensation secondary equipment set and a series compensation secondary equipment defect set which meet a preset statistical condition; dividing the operating age limit into a plurality of continuous operating age limit units based on the division granularity of the operating age limit; traversing the series compensation secondary equipment in the series compensation secondary equipment set, determining operation age limit units covered by the life cycle of the series compensation secondary equipment according to the commissioning time and the back operation time of the series compensation secondary equipment, and adding 1 to an equipment number counter corresponding to each covered operation age limit unit; traversing equipment defects in the series compensation secondary equipment defect set, determining an operation age limit unit to which the equipment defects belong according to the operation age limit of the corresponding series compensation secondary equipment when the defects occur, and adding 1 to a defect number counter corresponding to the operation age limit unit; and for each operating age limit unit, calculating the defect rate of the operating age limit unit according to the corresponding defect number and equipment number.

This invention discloses a method and apparatus for dynamic diagnosis of defects in high-altitude substations using cross-modal analysis. It belongs to the field of power operation and maintenance technology. The method includes acquiring monitoring data for each mode of the substation; solving the uncertainty evolution equation for each mode using the monitoring data to determine the uncertainty covariance matrix of each mode; determining the intrinsic connectivity strength between modes based on the similarity of the uncertainty covariance matrices between different modes; determining the geometric modulation attention weights between modes based on the intrinsic connectivity strength and the content relevance between corresponding modes; performing weighted fusion of each mode using the geometric modulation attention weights to determine the enhanced representation of each mode; and inputting the enhanced representation into a pre-trained defect diagnosis model to obtain the diagnostic results of the substation defects. By quantifying and fusing uncertainty in multimodal data to determine the enhanced representation for defect diagnosis, the accuracy of substation defect diagnosis is improved.

The invention discloses a health trend checking analysis method and device of switch cabinet equipment, a storage medium and equipment. The method comprises the following steps: acquiring real-time operation data, historical operation data, measuring point data, working condition data and simulation data of switch cabinet equipment; performing fusion analysis on the real-time operation data, the historical operation data and the simulation data by using a pre-trained prediction model to generate an equipment health state trend index; when an equipment early warning signal is monitored, early abnormal trend investigation analysis is carried out based on an equipment health state trend index, and an early warning and trend analysis result is obtained; and when an equipment fault signal is monitored, performing fault analysis on the historical operation data, the measuring point data and the working condition data based on a pre-constructed equipment defect knowledge base to obtain a fault analysis result. According to the method, analysis results of a physical mechanism and a data rule are fused, so that the method is more reliable than judgment simply depending on data or experience, and the false alarm rate and the missing report rate are greatly reduced.

The invention relates to a defect detection method and defect detection equipment. The defect detection method comprises the following steps: acquiring a standard TRS value S1; based on the standard TRS value S1 and a detection pressure value calculation formula, a detection pressure value F is obtained through calculation; applying detection pressure to the detection targets on the two supporting points according to the detection pressure value F, wherein a force application point is located between the two supporting points; wherein the calculation formula of the detected pressure value is F = (S1 * pi * d3) / (8 * L). Wherein d is the diameter of the target to be detected, and L is the distance between the two supporting points. According to the defect detection method, the transverse breaking strength is used as a standard for judging internal defects of the detection target, the standard TRS value S1 is converted into the detection pressure value F, and the force value is applied to the detection target. If the target is not fractured, it is indicated that the transverse fracture strength of the target can reach the standard TRS value S1, the target is a qualified product, the detection result is more reliable and stable compared with a drop detection method, and the probability of non-defective scrapping is lower.

The present disclosure relates to a device defect detection method, device, electronic device and computer readable medium. The method comprises: acquiring a real-time image of a device; inputting the real-time image into a first feature extraction model and a second feature extraction model to generate a feature extraction result; inputting the feature extraction result into a category prediction model to generate a target category; inputting the feature extraction result into a multi-scale prediction model to generate a plurality of target bounding boxes; and detecting defects of the device based on the target category and the plurality of target bounding boxes. The device defect detection method, device, electronic device and computer readable medium of the present disclosure have higher precision and faster detection speed compared with existing device defect detection technologies, and can be arranged in an edge service device.

The invention relates to the technical field of electric power system operation and maintenance, and particularly discloses an intelligent agent construction method and system for electric power inspection, which can automatically trigger an inspection task, schedule edge equipment to execute inspection and identify equipment defects based on a visual large model by collecting multi-source operation data of electric power equipment in real time and performing intelligent analysis. And generating a structured abnormal point location and a disposal suggestion, and finally forming a complete inspection report and pushing the inspection report to the operation and maintenance management platform. By automatically judging the risk and generating the targeted inspection task, manual intervention is not needed, and compared with a traditional mode depending on a fixed period or threshold alarm, task response is more timely, coverage is more accurate, the capacity of capturing sudden hidden dangers is effectively improved, intelligentization and automation of the whole inspection process are achieved, manual intervention is reduced, and the inspection efficiency is improved. The inspection efficiency and accuracy are improved; equipment abnormity and potential risks can be identified in time, and rapid disposal is supported.

The application discloses a kind of based on deep learning's concrete mixing station equipment visual defect automatic identification and grading system, belong to industrial automation detection technical field;The system includes image acquisition module, data preprocessing and enhancement module, deep learning defect identification module, defect severity intelligent grading module, result output and alarm module and system management and model updating platform;Image acquisition module obtains equipment surface image;Preprocessing and enhancement module optimizes processing to image;Deep learning defect identification module utilizes multi-task neural network model to identify equipment component and the type and position of its surface defect;Grading module is classified according to area proportion, geometric dimension etc. Quantitative index to defect severity;Output and alarm module generates detection report and triggers alarm.The application realizes the automatic detection and scientific grading of equipment defect, overcomes the disadvantages of manual inspection, and provides technical support for predictive maintenance.

The application provides an intelligent detection system of an RGB camera combined with a deep learning algorithm, comprising a defect feature adaptive engine, a hardware component, an algorithm component and a feedback component; the defect feature adaptive engine serves as a core control unit, links the hardware component, the algorithm component and the feedback component to realize full-link closed-loop detection; the defect feature adaptive engine performs the following operations: identifying scene working conditions and core working condition pain points, extracting the subdivided features of target defects in the scene and matching the feature label library, dynamically calling the adaptive hardware configuration, algorithm combination and preprocessing strategy, optimizing the detection parameters in real time, allocating the computing power resources according to the defect risk level and triggering the corresponding early warning, realizing the intelligent identification, positioning, quantification and grading early warning of defects of equipment such as pressure pipelines and pressure vessels, and through the implementation of the above system, the problems of poor adaptability, insufficient precision and unreasonable resource allocation of the existing system are solved, and the industrial detection is realized in a scene-based, refined and efficient manner.

This invention relates to the field of power equipment monitoring technology, specifically providing a training method and a defect identification method for a GIS equipment defect identification model. The training method includes: using source domain data and target domain data as training samples; extracting and fusing multimodal features from the training samples using a feature extraction module to obtain a fused feature vector; applying domain adversarial adaptation to the fused feature vector using a domain adversarial adaptation module to obtain the domain discrimination result of the training samples; using a prototype metric diagnosis module to measure the distance between the query set samples and multiple prototype centers in the training samples to obtain the defect category result of the query set samples; and determining the total loss function value based on the domain discrimination result and the defect category result to train the GIS equipment defect identification model. This invention solves the problems of scarce GIS field defect samples and domain offset between simulation data and measured data, achieving high-precision and strong generalization defect identification under complex working conditions.

The invention relates to an equipment defect detection method and system based on knowledge graph enhancement, and the method comprises the following steps: S1, obtaining the historical experience data of equipment, and constructing a knowledge graph through information extraction; s2, acquiring real-time monitoring data and generating a defect detection request; s3, analyzing the request intention, retrieving an associated path from the knowledge graph, and retrieving semantic similar fragments from the unstructured text; s4, embedding the real-time data, the structured query result and the semantic retrieval result into a unified vector space by adopting a graph neural network for feature fusion to form comprehensive context information; and S5, generating a defect diagnosis conclusion and a disposal suggestion based on the comprehensive context information. Through cooperative enhancement of the knowledge graph and the graph neural network, fusion of real-time monitoring data, historical experience and semantic information is realized, and the accuracy and real-time performance of equipment defect detection are improved.

The invention discloses a communication maintenance field operation big data management system based on visual analysis, and belongs to the technical field of detection and analysis. By separating illumination and reflection components, the detail definition of the target in strong light, backlight and dark scenes can be effectively improved, and target missing detection caused by uneven illumination is avoided; the high-frequency texture is reconstructed through the generative adversarial network, so that the detail recovery rate of the compressed video can be effectively improved, the problem of information loss in a compression-analysis link is solved, and data support can be provided for equipment defect identification; the extracted illumination feature sequence provides an environment context for subsequent risk early warning, and the false alarm rate is reduced; after YOLOv5s is combined with illumination features, the target detection precision in strong light and backlight scenes can be effectively improved, and the omission factor is reduced; the multi-modal feature matrix can be migrated to different maintenance scenes, the model weight is dynamically adjusted through the illumination features, and the robustness of the system is enhanced.

The invention discloses a transformer substation equipment surface defect detection method based on improved YOLO11, and belongs to the technical field of intelligent inspection of power equipment. The method comprises the following steps: constructing a substation equipment defect image data set, and performing data enhancement processing; a multi-scale efficient convolution module (EMSEC) is embedded in the YOLO11 backbone network, and the feature extraction capability is enhanced; a RepGDFPN is adopted to reconstruct a neck network, DySample is introduced for dynamic up-sampling, and the small target detection precision is improved; a detail enhanced shared convolution detection head (DESCHead) is designed, DEEConv and group normalization are fused, redundant calculation is reduced, and light weight of the model is realized; the trained model can be exported in an ONNX format and deployed on edge equipment, so that real-time defect detection is realized. According to the method, the model complexity is remarkably reduced while high detection precision is kept, and the method is suitable for edge equipment such as unmanned aerial vehicles and inspection robots and has wide industrial application prospects.