2089 results about "Data collecting" patented technology

The invention discloses a power distribution room anomaly detection system based on cloud side-end cooperation, and belongs to the technical field of intelligent power grids. In order to solve the problems of high network bandwidth pressure, insufficient edge computing capability, low anomaly detection accuracy, difficulty in multi-source data fusion and the like caused by the adoption of an end-cloud direct connection architecture in an existing power distribution room monitoring system, the system comprises: a data acquisition layer configured with various heterogeneous sensors to acquire operating parameters and environmental data in real time; the edge storage and calculation layer carries out local real-time processing, anomaly detection, model training and visual display, an anomaly detection module of the edge storage and calculation layer carries out research and judgment on real-time data to generate early warning information, and a prediction and detection linkage module monitors an anomaly probability trend and adjusts a sampling frequency; the edge gateway realizes protocol conversion and data forwarding; and the cloud decision-making layer aggregates multi-edge node data, optimizes a global model through federal learning, and issues and updates a local model. The system is used for improving the accuracy, real-time performance and reliability of anomaly detection of the power distribution room, reducing the operation and maintenance cost and realizing intelligent operation and maintenance.

Disclosed are a three-dimensional laser point cloud-based method and system for in-depth vegetation management in transmission corridors. The method comprises: collecting multi-source data by means of an unmanned aerial vehicle- and helicopter-mounted multi-sensor system, and preprocessing the multi-source data; using the processed multi-source data to create a digitalized power grid corridor and construct a vegetation management analysis model; on the basis of data analyzed by the vegetation management analysis model and geofencing technology, designing a hazard vegetation clearing-cost-compensation prediction model; and developing a mobile application for analyzing hazard vegetation data flows and the closed-loop management of vegetation management service flows. The described method achieves high-precision and high-coverage data collection, precisely identifies transmission line digital twins and hazard vegetation, standardizes and automates the management of hazard vegetation clearing and related compensation, and develops a mobile application for analyzing hazard vegetation data flows and the closed-loop management of vegetation management service flows, thereby achieving the end-to-end digital and mobile management of vegetation management. The organic combination of steps comprehensively improves the intelligence and efficiency of transmission line vegetation management.

The invention discloses a rubber product quality detection method and system, and relates to the related technical field of rubber quality detection.The method comprises the steps that rubber products are conveyed to a target detection table, and data acquisition of the rubber products is executed; inputting the wave impedance distribution data and the surface heat distribution gradient field into a defect causal atlas channel to execute phase inversion fusion processing, establishing a cross-modal coupling phase tensor, and identifying deep defect response characteristics by solving a phase difference matrix; activating a pressure spectrum sensing unit to apply control pressure, reading a local mechanical response spectrum, sending the local mechanical response spectrum to a cross-modal calibration layer, and executing joint error authentication; and performing quality detection report according to a self-calibration defect mapping result. The technical problems that in the prior art, multi-modal sensing data fusion efficiency is low, deep defects are difficult to comprehensively recognize, and consequently the quality detection precision and reliability of rubber products are insufficient are solved, and the technical effects that cross-modal cooperative detection is achieved, and the precision and reliability of defect recognition results are improved are achieved.

The invention relates to the technical field of water conservancy management, and discloses a water conservancy dynamic decision-making method and system based on digital twinning and multi-source heterogeneous data, and the method comprises the steps: carrying out the real-time information capturing of a target water conservancy system, and generating a system operation information scheduling matrix based on a grid geographic model and a time dislocation mark; performing hierarchical slicing on the system operation information scheduling matrix according to a multi-scale time window to form a multi-channel flow framework of regional hydrological dynamic behaviors; analyzing the evolution trends of the water level, the flow and the water quality by using a multi-channel flowing framework and a dynamic fusion analysis operator; dividing a river reach, a reservoir and a regulation and control facility corresponding to the potential risk event as candidate scheduling areas; and based on the local risk coupling network, dynamically optimizing a scheduling strategy, a pump gate operation sequence and a data acquisition frequency, and generating a real-time scheduling and emergency response scheme of the candidate region. The method has the advantage of improving flood early warning accuracy and water supply scheduling efficiency.

The invention discloses a cold region tunnel freeze injury diagnosis, risk grading and control system and method, and relates to the technical field of tunnel engineering. The system comprises a basic data layer for storing disease, index and measure databases; the core analysis layer is used for diagnosing a frost heaving mechanism, identifying single-factor, double-factor and three-factor frost heaving types, calculating freezing depth and frost heaving force and analyzing sensitivity; the risk level evaluation layer is used for dividing a disease zone, a frost heaving level, a freezing injury risk and a freezing injury risk response level; and the intelligent decision-making layer constructs an intelligent comprehensive decision-making analysis platform to realize risk-measure accurate matching and closed-loop management and control. The method comprises the four steps of data acquisition and storage, frost heaving mechanism diagnosis and quantification, risk grade evaluation and intelligent decision measure matching, through multi-source data fusion, multi-factor coupling analysis and zoning and grading prevention and control, the whole-process accurate management and control of the cold region tunnel frost damage is realized, and the frost damage treatment efficiency and the tunnel operation safety are improved.

The invention discloses an agricultural decision reasoning large model based on a large language model and a construction method thereof. The method comprises the following steps: constructing a space-ground-air integrated multi-modal agricultural data acquisition and fusion module, integrating data of a ground sensor, unmanned aerial vehicle remote sensing, meteorological sensing and Internet of Things, solving the problem of data isomerism through reinforcement learning and a Transform architecture, and ensuring real-time integrity of the data. In combination with crop simulation model mechanisms such as DSSAT and the like, the inference rule base is dynamically updated to realize rapid iteration of agricultural knowledge; an agricultural feature extraction module and an adaptation layer are integrated, so that professional data understanding is enhanced; a multi-stage training strategy is adopted, and large-scale agricultural text fine tuning is carried out; reinforcement learning is introduced and expert feedback is combined to optimize decision generation, and a causal inference module is innovatively added to improve the inference precision. The model fuses the language ability of a large language model and agricultural knowledge, can accurately predict and diagnose agricultural problems and provide decision suggestions, and significantly improves the intelligence, precision and high efficiency level of agricultural production.

The invention relates to the technical field of water supply pipeline detection, and discloses an underground water supply pipeline health grade evaluation and risk prediction method, which comprises the following steps: sensor arrangement: arranging a flow sensor, a pressure sensor and a sonic sensor at key positions of an underground water supply pipeline; and data acquisition: acquiring signals of the sensor in real time through a data acquisition module, wherein the signals comprise flow, pressure and sound wave signals. Preprocessing the data: carrying out preprocessing such as denoising and normalization on the collected signals; and multi-source data fusion: inputting flow, pressure and sound wave signals into a deep neural network model, and performing feature extraction and fusion analysis. And health level assessment: assessing the health level of the pipeline based on the output result of the deep neural network model. And risk prediction: predicting abnormal working conditions possibly occurring in the future and risk levels of the abnormal working conditions by analyzing the current pipeline state and historical data. And abnormal positioning: accurately positioning an abnormal position in combination with the propagation time of the sensor signal and a positioning model of the deep neural network.

The invention discloses an intelligent wheat yield prediction method and system, and relates to the technical field of agricultural information. According to the method, basic geography, climate, soil, crop physiology and agricultural management multi-source data are collected, and an input feature set is obtained through preprocessing and feature engineering; constructing an Attention-LSTM-CNN model fused with an improved attention mechanism, extracting local features through CNN, capturing time sequence association through LSTM, highlighting key contribution through an attention layer, and finishing model training in combination with an RMSE loss function and an Adam optimizer; and dynamically updating data and a prediction result in a wheat growth cycle, and outputting and triggering early warning in multiple forms. The system correspondingly comprises a data acquisition module, a data storage module, a data preprocessing module, a model calculation module, a prediction output and early warning module and a communication module. The method solves the problems that a traditional method is single in data, poor in model adaptability and lack of dynamic prediction, high-precision full-period prediction is achieved, and scientific support is provided for agricultural decision making.

The invention relates to an intrusion detection method based on cross-domain security management and a shared behavior model. The method comprises the following steps: acquiring multi-dimensional original data according to a preset cross-domain data acquisition rule and multi-domain node deployment; performing compliance, integrity and format matching degree verification on the original data, shielding sensitive information by using a dynamic desensitization technology based on a verification result, converting a heterogeneous data format, filtering missing field abnormal data, and obtaining compliance data; the method comprises the following steps: extracting a multi-dimensional feature set of user cross-domain access, constructing a shared behavior feature vector through weighted calculation, constructing a reference behavior model library in combination with a cross-domain security policy, and screening out intrusion behaviors and feature deviation data through feature comparison and behavior deviation calculation; according to the method, cross-domain security audit logs are fused for correlation analysis, intrusion behavior types and risk levels are judged by means of a Bayesian network model, differential security response strategies are generated and executed, and cross-domain intrusion detection and protection are achieved.

The invention relates to the technical field of environment intelligent supervision, and discloses a visual river and lake water area health ecological supervision and management system. A data acquisition and feature extraction module of the system is responsible for acquiring multi-source ecological monitoring data of water areas of rivers and lakes, extracting spatial change features of motion trails of aquatic organisms and a time sequence of water quality parameters from the data, and identifying abnormal features according to an index variation mode corresponding to a predefined ecological health level; and generating an ecological health abnormity response label set associated with the specific geographic position. And the health state analysis module obtains key information of the marked water area section based on the label set, analyzes a change track and a deviation interval of the health state in the supervision time frame, and determines a health state lagging time period. And the evaluation delay processing module is used for extracting related time information to calculate a time difference according to the water area subareas related to the lagging time period, screening out the water area subareas which are not evaluated completely and sorting the water area subareas, and generating an adjustable configuration scheme of the monitoring time period of the water area subareas.

The invention discloses a Beidou-based electric power tower attitude change on-line monitoring method, which comprises the following steps: A, setting a tower attitude acquisition system, and respectively acquiring multi-source data of a tower in real time by using the tower attitude acquisition system; b, constructing a meteorological database; c, constructing a tower attitude analysis model; the tower attitude analysis model comprises a data acquisition and preprocessing module, a multi-source data fusion module, a comprehensive analysis module, a situation diagnosis and risk assessment module, an intelligent prediction module, an early warning decision module and a visualization and feedback module; and D, on the basis of the trained constructed tower attitude analysis model, acquiring multi-source data of the tower in real time by using a tower attitude acquisition system, and performing online monitoring on the attitude change of the power tower in combination with a meteorological database. According to the invention, the posture change of the power tower can be obtained more accurately, and further prediction and early warning are carried out on the posture change of the power tower in combination with historical and predicted environmental data.

The invention discloses a data security protection method and system combined with big data analysis, and relates to the field of data security protection, and the method comprises the steps: building a multi-dimensional data collection and fusion data set according to API access records, external threat intelligence and context metadata association data; based on an isolated forest algorithm, calculating a sample path length to evaluate a behavior anomaly probability score; risk indexes of external threats, data sensitivity and permission exceptions are quantified respectively; constructing a data security risk comprehensive assessment model based on a weighted summation algorithm, and updating and optimizing the weight in the model by using a gradient boosting tree algorithm; and according to an output result of the data security risk comprehensive assessment model, setting a security risk level, and according to the security risk level, dynamically responding to a protection measure. The method has the advantages that continuous and dynamic risk assessment and automatic response to third-party data access behaviors are realized by fusing multi-source data and an intelligent algorithm.

The invention discloses a three-phase imbalance compensation method and system based on negative sequence current and a medium, and relates to the technical field related to power imbalance compensation, and the method comprises the steps: executing the three-phase current, voltage and reactive power data collection of a power distribution network, and carrying out the three-phase imbalance recognition through negative sequence current analysis; carrying out additional acquisition of a load mode and environment data on the power distribution network; taking the data acquisition result and the additional data set as input data to execute negative-sequence current change trend prediction; a double-layer compensation device is configured for cooperative control optimization; power distribution network balance compensation is executed through a time sequence compensation strategy, and time sequence window feedback is established; and carrying out adaptive strategy correction. The technical problems that in the prior art, a three-phase imbalance compensation strategy lacks perspectiveness and self-adaptive capacity and cannot cope with a dynamically-changing power grid scene, and consequently compensation precision and stability are poor are solved, and the technical effects that perspectiveness and predictive compensation is achieved, and three-phase imbalance compensation precision and stability are improved are achieved.

The invention provides an electric power system cross-source safety intelligent monitoring and dynamic response method and system based on mimicry defense, and relates to the technical field of electric power system network safety. According to the method, multi-source data are acquired by deploying an acquisition agent and a sensor and are encrypted and transmitted, parallel analysis and threat judgment are realized by utilizing a heterogeneous executor pool and dynamic scheduling, an attack chain is identified by combining cross-domain association modeling, disposal is performed according to grades, credible recovery is supported, and reinforcement learning and federated learning are assisted to optimize a strategy and resource allocation. And lightweight adaptation is realized through intensity control in a resource limited scene, a closed-loop safety system covering data acquisition, threat analysis, dynamic disposal and continuous evolution is formed, and the detection accuracy, response efficiency and robustness of a power system are improved.

Methods and systems for a monitoring system for data collection in an industrial environment including a data collector communicatively coupled to a plurality of input channels connected to data collection points operationally coupled to at least one industrial component in at least one of an industrial boiler system or industrial pipeline system; a data storage structured to store a library of stored noise patterns associated with operation of the at least one industrial component; a data acquisition circuit structured to interpret a plurality of detection values from the collected data; and a data analysis circuit structured to: analyze the collected data, determine a measured noise pattern for the at least one industrial component, and compare the measured noise pattern to the library of stored noise patterns to identify a changed condition of the at least one industrial component.

The invention discloses a coal drop pipe full-life-cycle operation and maintenance decision-making method. The method comprises the steps of S1, coal drop pipe full-life-cycle data collection; s2, constructing and updating a digital twinborn body of the coal drop pipe: S2.1, constructing a multi-dimensional digital twinborn body comprising a geometric model, a physical model, a behavior model and a rule model, and comprehensively describing the characteristics of each aspect of the coal drop pipe; s2.2, dynamically updating the twinborn body, and dynamically updating the twinborn body through a data driving algorithm by using real-time monitoring data; s3, health state evaluation and reliability prediction, including short-term health diagnosis and long-term reliability prediction; s4, carrying out dynamic operation and maintenance decision making based on risks and cost; and S5, performing decision issuing execution and closed-loop feedback. According to the method, the design data, the real-time monitoring data, the environment data and the operation and maintenance historical data are integrated, the coal drop pipe digital twinborn body is constructed and dynamically updated, it is ensured that the model is consistent with the physical entity state, and the comprehensiveness and timeliness of the data are improved.

The invention relates to an unmanned aerial vehicle electric power inspection monitoring method and system, and relates to the technical field of electric power inspection, and the method comprises the steps: obtaining multi-source data of a monitoring area, determining potential defect risk information based on the multi-source data, a power grid digital twin model and a fault propagation knowledge graph, and determining a potential defect priority based on the defect risk information, an inspection task is generated in combination with available unmanned aerial vehicle resources and an optimization algorithm, the inspection task is sent to an unmanned aerial vehicle cluster to collect data, so that the unmanned aerial vehicle cluster performs data collection on each to-be-verified risk area according to the unmanned aerial vehicle inspection task, and defects are identified based on the collected monitoring data and a defect identification AI model. And finally generating a maintenance decision suggestion based on the identification result and the risk information. The technical effects of accurately determining the potential defect risk of the power system, reasonably distributing the unmanned aerial vehicle resources for routing inspection, accurately identifying the defects and generating effective maintenance decision suggestions are achieved, and the efficiency and accuracy of power routing inspection monitoring are improved.

The invention relates to the technical field of robots, in particular to an instrument monitoring method, system and device for an inspection robot and a medium, the method is used for searching an instrument to be inspected by using the inspection robot and monitoring the instrument to be inspected, and the method comprises the following steps: obtaining a view scene image collected by the inspection robot, performing image segmentation and feature extraction on the view scene image; obtaining the exploration value of the instrument area; obtaining candidate paths from the inspection robot to the instrument areas; evaluating a cost function of each candidate path; carrying out adaptive visual angle adjustment and data acquisition on the inspection instrument; the monitoring method is an exploration type sensing strategy based on exploration value driving, and an instrument target can be found in the environment. And in combination with a historical task map, instrument semantic priori and current perception data, the information value of the candidate region is evaluated, and optimal decision making is performed on the path by fusing environmental factors such as illumination, shielding and equipment states, so that priority selection and dynamic reconstruction of the target region are realized.

The invention discloses an indoor water tank sediment pushing characteristic acquisition system and method, the system mainly comprises a water tank main body, a buffer unit, a drainage adjusting unit, a sediment throwing device, an imaging unit, an adjustable support and a data processing unit, and according to the method, non-contact data acquisition is carried out through a high-definition camera and a three-dimensional terrain measuring instrument; automatic identification, parameter extraction and migration speed calculation of landforms such as sand waves are achieved, the sediment bed surface is not touched in the whole process, damage to the bed surface form caused by traditional contact type measurement is eliminated, and the authenticity and reliability of data are ensured. Meanwhile, the buffer design of the combination of the flow slowing plate and the multi-stage water baffle is adopted, the kinetic energy of water flow is effectively dissipated through the graded water outlet structure, the impact force of injected water flow is greatly reduced, the bed surface is prevented from being scoured in the initial water injection stage, and the stability of the initial test condition is powerfully protected.

The invention discloses an enterprise carbon emission supervision and early warning system based on data collection, and the system comprises the following modules: a data collection and preprocessing module which is used for collecting the real-time state data of an enterprise carbon emission monitoring node, constructing a carbon emission time sequence, and executing the abnormality elimination to generate effective data; the carbon emission fitting calculation module is used for evaluating a multi-source data error index, distributing a weight and outputting a fitting value sequence; the emission offset detection module is used for recognizing a nonlinear severe offset event based on the fitting track; the trend prediction module is used for extracting trend components by adopting an improved Kolmogorov-Arnold network and generating a sudden change early warning mark; and the carbon emission risk early warning module is used for integrating the offset event and the trend mutation, generating a structured risk report and pushing the structured risk report to a supervision system. According to the invention, the carbon emission abnormity identification precision and the early warning response efficiency are improved, and intelligent perception and linkage regulation and control of enterprise carbon emission risks are realized.

The invention discloses an air-space integrated safety monitoring method for distribution network construction based on an unmanned aerial vehicle, and the method comprises the following steps: A, constructing an air-space integrated cooperative monitoring network, and collecting the multi-source data of a distribution network construction region through the air-space integrated cooperative monitoring network; c, constructing a construction safety analysis model; the construction safety analysis model comprises a data acquisition and preprocessing module, a multi-source data fusion analysis module, a risk diagnosis and evaluation module, an intelligent prediction and early warning module, an early warning decision and disposal module and a visualization and feedback module; and D, based on the trained construction safety analysis model, collecting multi-source data by using an air-space integrated cooperative monitoring network, and carrying out online monitoring, early warning and disposal on the construction safety of the distribution network in combination with a meteorological database. Unmanned aerial vehicle near-earth monitoring and satellite remote sensing data can be fused, the distribution network construction process is effectively monitored, and potential safety hazards are reduced.

The invention discloses an engineering vehicle safety simulation and prediction system based on digital twinning, and the system comprises a data collection layer which collects multi-source heterogeneous data in real time; in the knowledge graph layer, a streaming inference engine is constructed based on an Apache Jena graph database, and an entity-relationship-attribute triple dynamic graph structure is adopted; according to the AI model layer, a physical rule serves as a loss function constraint term to be embedded into a neural network through a physical information neural network, a digital organ model concept is combined to split a vehicle into key organs for heterogeneous modeling, a simplified physical model is adopted in the core physical process, and an LSTM-AI model is adopted in external behaviors; the explanatory analysis layer is used for integrating an SHAP / LIME explanatory tool to output a visual evidence chain during fault prediction, and deploying an online incremental learning framework to allow the model to learn from new data and dynamically adjust normal range definition; and the visualization and application layer is used for performing three-dimensional visualization rendering based on WebGL or Three.js, and ensuring data transmission security through block chain evidence storage and end-to-end encryption.

The invention provides a multi-stage desensitization electricity charge sensitive data protection system, belongs to the technical field of information security in the power industry, and relates to a full-life-cycle protection scheme of sensitive data such as an identity label, account information and contact information in an electricity charge service system. The problems that in the prior art, only simple masking or full-table encryption is carried out on a database layer, the desensitization dimension is single, multiple links are prone to leakage, no role fine-grained desensitization strategy exists, and exported files are difficult to trace are solved. The system comprises a data acquisition source and acquisition layer desensitization, a transmission layer encryption, a storage layer transparent encryption, an API gateway, a desensitization strategy center, a front-end dynamic mask, a watermark export, a log desensitization plug-in and an auditing log module, covers the whole process from data acquisition to log auditing, dynamically regulates and controls data display according to roles, and exports a file embedded identity watermark. And desensitizing the log and reserving operation marks. Full-link leakage prevention, fine-grained permission control and leakage traceability can be realized, and the transformation cost is low.

The invention discloses an action-oriented teaching model construction system based on an AI large model, and relates to the technical field of big data analysis, and the system comprises a student ability portrait module which is used for generating student ability portraits and recognizing the advantages and weak links of students through multi-dimensional data collection and AI large model analysis; and the typical task intelligent generation engine automatically generates typical tasks according with student ability levels based on the student ability portraits and the teaching outlines. According to the method, multi-dimensional data such as historical homework and test scores of students are integrated, deep analysis is carried out by using an AI large model, and accurate student ability portraits are generated, so that teachers are helped to comprehensively understand the knowledge mastering degree and the skill proficiency of the students, the advantages and weak links of the students can be accurately identified, and the students' ability and proficiency of the students are improved. Therefore, the typical tasks conforming to the ability level of the students are automatically generated, and the teaching content is ensured to be matched with the actual demands of the students, so that real personalized teaching is realized.

The invention discloses a comprehensive remote sensing recognition system for hidden danger of ice lake outburst. The system comprises a data acquisition and preprocessing module, an ice lake extraction module, an ice dam stability evaluation module, a risk calculation module and a visualization and early warning module. According to the system, optical images, SAR images, LiDAR point cloud and meteorological data are utilized, firstly, a glacial lake boundary is extracted through a water body index, the area and the volume are calculated, then the water level rising rate is monitored in combination with multi-temporal data, the dam body stability is evaluated by utilizing three-dimensional terrain and deformation parameters, and the dam body stability is evaluated by combining the temperature melting corrosion rate and triggering factors such as rainfall, earthquakes and inflow. And constructing a multi-factor outburst risk index model, and finally carrying out visualization and graded early warning on a result in a GIS platform. The method can achieve the quick, comprehensive and intelligent recognition of the hidden danger of ice lake outburst, has the advantages of being high in monitoring precision, high in automation degree, high in early warning timeliness and the like compared with a traditional single data or experience judgment method, and can be widely applied to the disaster prevention and control work of dense ice lake areas such as the Qinghai-Tibet Plateau and the Himalaya mountain.

The invention relates to the technical field of geological disaster monitoring and early warning, and discloses a geological disaster early warning system based on multi-source data fusion. The system comprises a data acquisition and preprocessing module, a dynamic coupling modeling module, a space mapping and feature recognition module and a risk analysis and early warning generation module. Firstly, transient disturbance and long-term steady-state components in monitoring data are separated; a dynamic coupling model containing bidirectional geomechanical feedback is constructed, the component fusion proportion is automatically adjusted according to the feedback intensity, and physically consistent fusion data is generated; spatial mapping is carried out by using an adaptive grid, and an effective abnormal feature cluster is identified through parallel scanning and prior geological knowledge constraint; and constructing a causal graph based on the abnormal clusters to carry out risk assessment and early warning. According to the system, physical driving of data fusion and knowledge guidance of anomaly recognition are realized, and the accuracy and reliability of early warning are improved.

The invention discloses an air-ground collaborative observation unmanned system key target intelligent search method, which comprises the following steps that: an unmanned aerial vehicle carries a positioning module to provide a positioning coordinate, and executes a preset fixed-point cruise task; the unmanned aerial vehicle performs data acquisition on the environment of the cruising area; a height self-adaptive target detection module is adopted to carry out target detection on image data collected by the unmanned aerial vehicle, a rescue target and temporary obstacles generated after a disaster are recognized, coordinate transformation is carried out on the obstacles and target points, and a detection result is mapped to a position under a global coordinate system; transmitting the recognized target point and obstacle coordinates to a ground mechanical dog in real time through a communication module; the ground mechanical dog updates a dynamic cost map according to the received environment information to realize real-time perception of the environment; and the ground mechanical dog performs path planning based on the environment map. According to the method, efficient perception and rapid rescue of the target point can be realized, and the task execution efficiency and the system robustness in the post-disaster complex environment are remarkably improved.

The invention belongs to the technical field of computer vision, and discloses a target identification tracking method based on a self-supervision mechanism, and the method comprises the steps: enhancing a self-supervision pre-training module through causality, constructing a causal sample pair through unlabeled video data, learning universal features through combining with comparison loss, and achieving the high-precision tracking without large-scale manual labeling. A multi-modal feature fusion and dynamic calibration mechanism further reduces dependence on annotated data, is especially suitable for industrial inspection, field monitoring and other scenes where data acquisition is difficult, significantly reduces time and labor costs in a data preparation stage, and broadens the application range of the technology in resource limited scenes; a causal reasoning and physical constraint mechanism is introduced, a dynamic relation between targets is modeled through a space-time causal graph, unreasonable tracks are filtered in combination with a physical rule, and complex conditions such as shielding, rapid movement and extreme weather are effectively dealt with; the dynamic feature calibration module corrects feature drift in real time, and ensures stable model performance in long-term tracking.

The invention relates to the technical field of power system state monitoring and operation and maintenance, and discloses a power distribution network equipment state intelligent sensing and predictive maintenance decision-making method, system, equipment and medium, and the method comprises the steps: collecting equipment operation data, monitoring data and environment data through a power distribution automation system, an online monitoring device, an environment sensor and other channels; performing time synchronization and quality verification on the multi-source data; extracting multi-dimensional characteristic parameters reflecting the running state of the equipment, and performing normalization processing and fusion analysis on the characteristics; establishing an equipment state evaluation model by using a deep learning algorithm, and carrying out quantitative scoring on the current health condition of the equipment; constructing a time sequence prediction model, and predicting the future degradation trend and residual life of the equipment; and establishing a multi-objective optimization model, and generating an optimal maintenance decision scheme. The whole process is from data acquisition, state perception and fault prediction to decision optimization, and intelligent management of power distribution network equipment is realized.

The invention discloses a welding quality on-line detection method based on machine vision, and particularly relates to the field of welding quality detection.The welding quality on-line detection method comprises the steps that multi-dimensional basic data collection is conducted, the two-dimensional form, heat distribution, micro texture, three-dimensional topological data and welding process parameters of a welding seam are obtained, and a full-dimensional original data set with time-space synchronization is constructed; performing feature decoupling on the data, respectively constructing geometric morphology, thermal, texture and three-dimensional topological feature functions, and converting unstructured data into structured feature values; integrating multi-dimensional features through a nonlinear deep fusion model, introducing thermal probability mapping, texture attention weighting and a three-dimensional residual compensation mechanism, and outputting a weld comprehensive quality value; and finally, positioning dominant defect dimensions and identifying defect types in combination with a dynamic threshold model to finish quality grading judgment. Multi-dimensional accurate detection and dynamic grading of the welding quality are achieved, different scene requirements are met, and the detection efficiency and the judgment accuracy are improved.