1712 results about "Incremental learning" patented technology

The invention relates to a risk management and control method and system based on real-time behavior analysis, and the method comprises the steps: carrying out the structural processing of multi-source behavior data through lightweight protocol decoding and behavior label embedding, and constructing an original behavior data set of a user and an entity; extracting multi-dimensional behavior characteristics by using a sliding window analysis and sparse representation mechanism, and constructing a user behavior graph by combining graph embedding learning; constructing a time-sensitive behavior trend model through streaming modeling and an incremental learning strategy, identifying an abnormal evolution trajectory in real time, and introducing a dynamic risk threshold regulation and control mechanism; adopting a high-throughput flow data processing and fast similarity matching algorithm to construct a fusion discrimination model, giving risk levels to abnormal behaviors and classifying the abnormal behaviors; and finally, performing closed-loop optimization in combination with a historical treatment effect. The system has the advantages of high real-time performance, high calculation efficiency, adaptability to complex network environments and the like, and the network security protection capability can be effectively improved.

The invention discloses a power grid load prediction and scheduling optimization system based on artificial intelligence, particularly relates to the technical field of power system automation, and solves the technical problems of low power grid load prediction precision, poor scheduling strategy robustness and insufficient source grid load storage coordination in the prior art. Multi-source heterogeneous data space-time alignment is realized by constructing a data acquisition layer based on edge calculation, a load prediction result is generated by adopting an AI prediction module fused by a graph convolutional network and an attention mechanism, and a source-network-load-storage collaborative scheduling scheme is generated through a multi-target risk hedging optimization algorithm. And closed-loop optimization is realized by using digital twinborn pre-check and incremental learning. And finally, the load prediction accuracy, the scheduling decision reliability and the system adaptive capability in the new energy access environment are improved.

The invention relates to the technical field of agricultural information management, and particularly discloses an agricultural information management system and method based on a big data platform, and the method comprises the steps: firstly deploying a multi-source data collection module at an edge calculation node, and obtaining and standardizing the soil moisture content, meteorological environment and equipment operation data in real time; secondly, constructing a local dynamic irrigation strategy model, and realizing multi-objective optimization through a reinforcement learning algorithm; establishing a federated learning framework at the cloud, dynamically distributing node weights by adopting an attention mechanism, and realizing model aggregation of privacy protection in combination with secure multi-party computing; an optimal irrigation instruction is generated through a multi-source data fusion engine, and a three-level response exception handling mechanism is established; and finally, a closed-loop feedback system containing short-term incremental learning and long-term architecture optimization is formed. The corresponding management system comprises six functional modules, namely a data acquisition module, a local modeling module, a federated learning module, a real-time decision-making module, an abnormal monitoring module and a closed-loop optimization module.

The invention relates to the technical field of elevator equipment state monitoring and fault diagnosis, and discloses an elevator traction machine running state real-time diagnosis method and system based on high-frequency sampling. According to the method, vibration (larger than or equal to 20 kHz), current (larger than or equal to 10 kHz), sound / sound emission, temperature and rotating speed signals of a traction machine are synchronously collected through a high-frequency multi-mode sensor array; capturing early weak fault transient characteristics; the edge computing unit completes data preprocessing, time synchronization, feature extraction and anomaly detection, and uploads key data to a cloud end through cloud-edge collaboration; the cloud end adopts a working condition self-adaptive strategy and a multi-modal fusion model to carry out deep diagnosis, and outputs fault types, positions and grades; and combining incremental learning and a degradation model to realize health quantification and residual life prediction. Through fusion of high-frequency data capture and an intelligent algorithm, the early fault detection capability, variable working condition adaptability and diagnosis real-time performance of the traction machine are improved, and a solution is provided for predictive maintenance of an elevator.

The invention discloses a fault root cause positioning method and system driven by a dynamic knowledge graph, and relates to the technical field of fault root cause localization, and the method comprises the steps: collecting and obtaining a multi-source fault associated data set, carrying out the entity association extraction of the multi-source fault associated data set, and obtaining a fault entity set and an entity relationship set; performing graph node cascading and incremental learning updating, and constructing a fault updating knowledge graph; monitoring and acquiring target fault data, performing mode matching reasoning, and generating a fault mode candidate root cause set; and performing similarity matching on the fault mode candidate root cause set in combination with a historical fault case library, and determining a target fault root cause positioning result. The technical problem of low fault diagnosis efficiency caused by inaccurate fault root cause positioning and knowledge graph updating lagging in the prior art is solved, and the technical effects of realizing accurate positioning of the fault root cause and dynamic improvement of the knowledge graph and improving the fault diagnosis efficiency and accuracy are achieved.

The invention discloses a numerical control machine tool machining environment monitoring system, and belongs to the technical field of intelligent monitoring. Comprising the following modules: a multi-dimensional sensing monitoring module for realizing omnibearing data acquisition of an interaction state of a cutter and a workpiece in a machining process; the feature extraction module is used for converting the time sequence data into a key feature set for representing the state of the cutter; the wear type identification module is used for accurately identifying and classifying the wear type of the cutter based on a deep learning classifier and judging the specific wear type; the wear progress prediction module calls a corresponding special prediction model according to the identified specific wear type, quantifies the wear progress rate and estimates the residual life of the cutter; the decision support module is used for integrating the tool wear state and the prediction result, balancing the production efficiency, the machining quality and the tool cost, and providing optimization suggestions of parameter adjustment and tool changing opportunities; and the self-learning optimization module is used for continuously collecting actual production data to carry out model evaluation and incremental learning so as to realize multi-machine knowledge sharing.

The invention, which relates to the technical field of payment security, discloses a payment-scene-oriented interaction intention identification and error correction system comprising an input analysis module, an intention simulation module, a dynamic decision module, a biological verification module, an audit evidence storage module, and a cross-scene knowledge migration module. According to the method, multi-modal data such as voice, texts, images and touch tracks are integrated, structured feature vectors are generated through a cross-modal attention network, the problem of incomplete single-modal coverage is solved, cross-modal data consistency verification is achieved based on a unified semantic tag system, and the reliability of input sources is graded by combining equipment fingerprints and geographic positions, so that the reliability of the input sources is improved. A high-risk transaction protection capability is enhanced, a generative adversarial network is utilized to construct a virtual attack sample library, attacks such as tampering with characters similar in shape and AI faking voiceprints are simulated, unknown threats are actively defended through cosine similarity matching, a user historical behavior statistical model is integrated, and known risks such as high-frequency small-amount transfer are passively intercepted. And a closed-loop incremental learning continuous optimization model is supported.

The invention discloses a reinforcement learning-based virtual power plant response optimization scheduling system and method, and relates to the technical field of virtual power plant intelligent scheduling. The system comprises an environment modeling module, an intelligent agent module, a multi-agent coordination module and a self-adaptive optimization module which are respectively used for constructing a multi-dimensional state space and a layered action space, generating and optimizing an action strategy based on an Actor-Critic network, executing a scheduling instruction through a layered multi-agent structure and realizing conflict consensus, and dynamically adapting to state space change in combination with incremental learning and meta-learning mechanisms. The system and the method have the advantages of fine state modeling, efficient action response, adaptive strategy updating, stable agent coordination and the like, and can keep the continuity, the stability and the optimality of a scheduling strategy in an operation environment in which multi-source heterogeneous power resources participate in scheduling cooperatively, market rules change frequently and load fluctuation is violent.

The invention discloses a multi-modal dynamic fusion and incremental learning fault diagnosis method for deep vertical shaft equipment, which belongs to the technical field of industrial equipment fault diagnosis, and comprises the following four steps of: constructing a pre-training large model to perform feature extraction, and relying on a multi-layer Transformer encoder and a dual loss function, establishing a multi-modal dynamic fusion and incremental learning fault diagnosis model; mining cross-modal universal fault features from vibration, temperature and current multi-modal time sequence data; according to the method, multi-modal features are fused, multi-modal association is constructed, modal weights are dynamically adjusted through a modal gating unit and a time delay compensation attention mechanism to adapt to signal quality changes, and meanwhile time sequence deviation is corrected to achieve accurate association; incremental learning is realized by using a decoupling projection layer, and a lightweight projection module is designed for a newly added fault task to suppress disastrous forgetting; network training is optimized, pre-training loss, incremental learning loss and attention regularization loss are integrated through a multi-objective loss function, and model stability and diagnosis precision are improved. The method has the advantage that the model stability and the diagnosis precision are improved.

According to the multi-modal automatic knowledge graph construction method based on the large language model, a multi-modal data stream is preprocessed, features are extracted, and the multi-modal data stream is mapped to a unified semantic space through a cross-modal alignment network after being processed through the large language model, a visual converter and a time sequence neural network. In the space, entities and categories are recognized based on a large language model, a triple is generated by combining a multi-modal feature judgment entity relationship, mapping fusion is performed through an ontology alignment algorithm driven by a graph neural network and a predefined domain ontology, finally knowledge is stored in a graph database, and dynamic updating is performed by means of incremental learning and online reasoning. Standardized APIs and visualization components are provided. According to the method, the construction efficiency and the automation degree of the knowledge graph are remarkably improved, the cross-modal information fusion and knowledge maintenance capability is enhanced, and the application requirements of intelligent retrieval, recommendation, decision support and the like are met.

The invention provides an intelligent cooperative control method and system for a multi-mode phototherapy device, and the method comprises the steps: obtaining physiological parameters of a plurality of users to form a data set, and constructing a spectrum feedback model in combination with historical phototherapy data and environment data; on the basis of the model and historical phototherapy target data, a multi-mode spectrum parameter model is generated through a preset neural network, and cooperative control over multi-light-source equipment is achieved; physiological parameter changes of a target user are monitored in real time, and monitoring data are sent to an edge computing node; the edge node dynamically adjusts a spectral parameter combination scheme according to the monitoring data and a safety threshold, and generates a treatment effect evaluation index; and establishing a user spectrum feedback feature library based on the evaluation indexes, and continuously optimizing a spectrum feedback model through incremental learning. According to the invention, intelligent cooperative control of the multi-mode phototherapy equipment can be realized, the energy utilization efficiency of the phototherapy equipment is improved, the spectrum feedback difference influence among different users is reduced, and the accuracy and safety of the phototherapy effect are enhanced.

The invention belongs to the technical field of financial data management, and particularly discloses a financial multi-source protocol self-adaptive fusion system based on AI semantic understanding and a knowledge graph, which comprises the steps of avoiding fusion errors caused by semantic misunderstanding through deep semantic analysis and a conflict resolution decision based on the knowledge graph; when protocol version updating is detected, incremental learning and model adjustment are carried out based on a newly added sample and historical experience, and a fusion protocol standard is dynamically updated, so that the high adaptive capacity to dynamic change of a financial protocol is realized; an exception monitoring and repairing mechanism is introduced, data missing, format errors and other exceptions are found in time and repaired online based on historical data modes and business logic, and negative influences of abnormal data on downstream risk control, transaction decision making and other key business links are avoided; through deep collaboration and information feedback among the intelligent modules, an intelligent system capable of self-learning and evolution is constructed.

The invention discloses a territorial space planning dynamic monitoring method based on multi-source data fusion, and belongs to the technical field of territorial space planning intelligent monitoring, and the method comprises the steps: obtaining data, and generating a multi-source heterogeneous data set; performing space-time alignment processing by using a preset regional association rule of a planning knowledge base to generate a space-time unified data set; constructing a dynamic knowledge graph taking planning elements as a core based on the data set, and updating node relation weights in real time; guiding a multi-source data fusion direction through the map relation weight to generate a fusion feature vector; incremental learning monitoring processing is carried out on the feature vectors, parameters are optimized, and a planning implementation state monitoring result is output; and updating the knowledge graph node relation weight in a closed loop manner according to a monitoring result, and synchronously optimizing incremental learning monitoring processing. According to the method, a dynamic knowledge graph is adopted to guide data fusion and an incremental learning closed-loop optimization mechanism in real time, and accurate perception and adaptive decision support of a planning implementation state can be realized.

The invention discloses an energy storage power supply residual life prediction method and system based on dynamic weight distribution, and belongs to the technical field of energy storage power supply health management. The method comprises the following steps: acquiring voltage, current and temperature time sequence signals of energy storage power supply operation through a multi-source sensor, and constructing a degradation characteristic sequence by adopting a sliding window method; a degradation inflection point division stage is detected and identified by using a curvature extreme value, a historical degradation mode is matched based on a dynamic time warping algorithm, and an optimal model group is selected; respectively carrying out dynamic weight distribution on the time step length and the feature dimension by adopting a dual-channel attention network, and fusing a learnable coefficient with space-time attention output to generate a life prediction result; residual errors and feature drift are monitored and predicted in combination with an incremental learning mechanism, and model parameters are updated and a degradation knowledge base is expanded by adopting an elastic weight consolidation algorithm. According to the method, the key information capturing capability is enhanced through a space-time attention mechanism, the model adaptability is optimized in combination with incremental learning, and the energy storage power supply life prediction precision and the working condition generalization performance are remarkably improved.

The invention relates to the technical field of artificial intelligence, in particular to a real-time anti-fraud monitoring system and method based on behavior reasoning and sentiment analysis, and the system comprises a multi-modal data collection unit, an edge preprocessing unit, a feature fusion and behavior reasoning unit, a large language model context reasoning unit, a risk assessment and decision unit, and an intervention execution unit. A log recording and federal incremental learning unit; the method has the beneficial effects that the traditional isolated single-mode detection is evolved into an emotion and behavior dual-channel collaborative multi-mode recognition system through millisecond-level coaxial alignment of voice, video and user operation logs; the robustness of dialect, noise and expression shielding is greatly improved through the multi-modal fusion model, so that the cross-scene recognition accuracy is improved by nearly three percent compared with that of a traditional single-voice scheme, and high-sensitivity capture of hidden and emotion control type fraud is truly achieved.

The invention relates to the technical field of risk analysis, in particular to a supply chain risk quantitative evaluation method and system based on a dynamic affair atlas, and the method comprises the steps: collecting multi-source heterogeneous data, constructing a four-dimensional space-time model comprising a time dimension, a geographic space dimension, a supply chain network space dimension and a risk influence space dimension, representing the supply chain event as four-dimensional spatio-temporal data; a supply chain entity is identified from the four-dimensional spatio-temporal data, risk events are extracted, a affair graph is constructed, and the affair graph takes the risk events as nodes and the evolution relation between the events as edges to calculate the relation weight between the events; calculating a probability quantized value of the risk conduction path based on the affair map, and obtaining a comprehensive risk score of the target entity; generating a risk mitigation strategy based on the comprehensive risk score; and monitoring the deviation between the actual risk occurrence condition and the prediction result, and updating the affair map and the risk mitigation strategy through adaptive parameter optimization and an incremental learning mechanism to form a self-evolutionary risk assessment system.

The invention discloses an industrial image anomaly detection method based on deep learning, and particularly relates to the technical field of industrial visual detection. The problems of high false alarm rate, fuzzy fine defect positioning, insufficient real-time response capability, difficulty in model increment updating and the like caused by data distribution drift in an industrial scene are solved. According to the method, robust features are extracted through a multi-scale feature fusion auto-encoder, and a dynamic memory bank is constructed to update a normal sample prototype online; a dual-path detection mechanism is adopted to cooperate with a pixel-level reconstruction error and attention weighted feature matching deviation; efficient edge reasoning is realized in combination with block parallel processing and model compiling optimization; and designing an elastic incremental learning framework to prevent disastrous forgetting. And finally, false alarms caused by environmental changes are reduced, accurate positioning of pixel-level defects is realized, millisecond-level detection requirements of high-resolution images are met, safe and efficient model online evolution is supported, and adaptability and reliability of an industrial quality inspection system are comprehensively improved.

The invention discloses a method and a system for accessing and registering power Internet of Things equipment in a multi-protocol environment. The method comprises an original communication protocol identification step, a communication data semantic identification step, a target communication protocol conversion step and a device registration step. According to the invention, protocol identification, dynamic protocol conversion and equipment unique ID registration are carried out on equipment accessed to the electric power Internet of Things for the first time, intelligent identification and semantic level conversion of heterogeneous protocols are realized, incremental learning of unknown protocols is supported, real-time communication between equipment and a platform and between equipment is supported, and the real-time communication between equipment and platform is realized. And the compatibility and the adaptive capability of the electric power Internet of Things system are improved.

The invention relates to the technical field of motor state monitoring and fault prediction, and provides a motor residual life analysis method and system based on a support vector machine, and the method comprises the steps: building a multi-dimensional degradation feature set of a motor, and calculating a health index degradation rate based on multi-source sensor data and a failure threshold; constructing a support vector machine regression model of an adaptive kernel function, predicting a health index change track by using the model and real-time data, and generating a residual life evaluation result if a prediction deviation is within an allowable error range; otherwise, starting an incremental learning mechanism to update the training data set, dynamically adjusting kernel function parameters, and recalculating the trajectory; and if the error requirement is still not met, model regularization parameters are optimized in combination with the working condition data until the residual life evaluation result is converged. According to the method, the accuracy and dynamic adaptability of motor residual life prediction can be improved, and the robustness of the model to complex working conditions is enhanced.

The invention relates to the technical field of science and technology operation management and control, and discloses a science and technology operation intelligent management and control method and system based on big data. According to the method, firstly, heterogeneous data sources in the scientific and technological operation process are collected, and a standardized operation data set is generated through multi-modal fusion processing; performing dynamic feature classification on the key operation indexes, extracting time sequence features and spatial correlation features of the key operation indexes, and constructing a multi-level operation state graph based on feature importance weights; then matching a service rule base with the atlas, identifying abnormal nodes and resource conflict paths, and generating an optimization instruction set containing node repair priorities and conflict resolution strategies; an executable management and control operation sequence is generated; and finally, collecting a feedback data stream, and updating the service rule base and the feature importance weight through an incremental learning mechanism to form a closed-loop optimization link. According to the method, heterogeneous data can be effectively integrated, the operation problem can be accurately identified, the optimization strategy can be quickly generated, and intelligent management and control and continuous optimization of scientific and technological operation can be realized.

The invention discloses a preschool education field knowledge base construction method fusing a knowledge graph and a large language model, and relates to the technical field of preschool education method optimization. Multi-channel collection of preschool education data and cleaning preprocessing; utilizing a customized model to extract entities and relationships, and constructing a structured knowledge graph; performing fine tuning on the big language model in the preschool education field; fusing atlas and model knowledge through a gating mechanism, and performing dual verification; establishing a monthly updating mechanism to realize incremental learning; in addition, personalized knowledge recommendation is realized based on a user portrait, interactive questions and answers are optimized through combination of retrieval and a model, and data security is guaranteed by adopting anonymization and authority control. Through multi-source data integration and intelligent technology fusion, an accurate and dynamic preschool education knowledge base is constructed, requirements of children are more accurately understood, content fitting cognition is generated, personalized learning recommendation is realized, interaction experience and knowledge service quality are improved, data security is guaranteed, and digital upgrading of the preschool education field is promoted.

The invention relates to the field of recommendation systems, and particularly discloses a personalized recommendation method based on multi-modal behavior sequence modeling, which comprises the following steps of: acquiring multi-modal behavior data such as user text, image, time and place, preprocessing, and realizing dynamic fusion of the data by utilizing a multi-modal self-attention mechanism (MMSA) to obtain a multi-modal behavior sequence model; and the interest evolution of the user is accurately captured. An independent RNN module is adopted to model long-term and short-term interests of a user, and the long-term and short-term interests are combined through a self-learning weight coefficient, so that the change of the user interests is reflected more accurately. In addition, by introducing an online learning and incremental learning mechanism, model parameters are dynamically adjusted according to real-time feedback of the user, and it is ensured that a recommendation result can respond to user interest changes in time. According to the method, the defects of an existing recommendation system in the aspects of data fusion, time sequence modeling and real-time adaptability are effectively overcome, recommendation individuation and accuracy are improved, and the real-time updating capacity and scene adaptability of the system are enhanced.

The invention provides a shield tunnel dynamic settlement compensation construction method based on an adaptive optimization algorithm, and the method comprises the steps: collecting the ground surface settlement, soil stress and underground water level data in real time through an Internet of Things sensor, achieving the data preprocessing and feature extraction in combination with an edge calculation node, constructing a three-dimensional geologic model, integrating the historical engineering data through transfer learning, and achieving the dynamic settlement compensation of a shield tunnel. The method comprises the following steps: identifying a high-risk area by using a clustering algorithm, designing a hybrid adaptive optimization framework with fusion of random forest and incremental learning, dynamically adjusting shield tunneling speed and soil bin pressure construction parameters, introducing an adaptive step length mechanism to cope with geological complexity change, and identifying a settlement abnormal mode through Fourier transform. Precise compensation is achieved in combination with a layered grouting strategy, the pressure of a soil bin is dynamically adjusted based on a hydraulic system, a closed-loop feedback mechanism is established, the predicted deviation rate is compared with an actual monitoring value, model parameters and the compensation strategy are continuously optimized, the settlement control precision is improved, and the construction risk is reduced.

The invention discloses an industrial time sequence prediction method based on adaptive continuous learning. The method comprises the following steps: firstly, dividing a non-stationary industrial time series data set to obtain a plurality of domains with the maximum distribution difference; different time domains are then modeled in sequence, and an improved empirical playback (DER + +) method is used to avoid catastrophic forgetting of previously accumulated knowledge. Meanwhile, a soft sample buffer area is introduced to promote memory and learning of key modes in the current field. And finally, the time-sensitive activation function TimeRelu enables the time convolutional network (TCN) to have a time evolution property, and the generalization ability of the prediction model is enhanced. According to the method, the continuous learning normal form is introduced into the time sequence prediction task, the limitations of huge resource overhead of traditional cumulative training, disastrous forgetting of an incremental learning mode and the like are overcome, and the method has theoretical and practical significance on industrial time sequence prediction.

The invention discloses a target intelligent cooperative identification method based on an unmanned aerial vehicle cluster, and belongs to the field of unmanned aerial vehicle cluster control and computer vision. According to the method, cluster networking and model initialization are realized through a dynamic heterogeneous federated learning architecture; a space-time attention mechanism is adopted to optimize task allocation, and a deformable network is utilized to extract multi-view target features; a cascade characteristic distillation fusion strategy is provided, and modal compression and cross-modal gating fusion are carried out on multi-source data such as multispectral data and laser radar data; an anti-interference elastic communication mechanism based on meta-learning is designed, and the system robustness is enhanced by combining space-time confrontation detection and a dynamic spectrum sensing technology; an unsupervised federal incremental learning system is established, and online evolution of the model is realized through momentum weighted aggregation. According to the method, the target identification accuracy is improved by 35% in a complex environment, the time delay is reduced to 200 ms, and high-precision real-time identification support is provided for military reconnaissance, disaster rescue and other scenes.

The invention discloses a wind turbine generator performance dynamic evaluation method and system based on multi-source data fusion, and the method comprises the steps: synchronizing multi-modal heterogeneous data through a quantum encryption algorithm and an edge gateway; constructing a space-time semantic graph network through a deep semantic analysis technology, and generating a precise space-time feature matrix; based on an attention mechanism, generating a physical enhancement feature vector; constructing cross-working-condition health index mapping by applying a machine learning algorithm, and quantifying a cross-working-condition comparable health index; a hierarchical incremental learning architecture is adopted, and a multi-objective optimization algorithm is utilized to generate a dynamic maintenance priority sequence; and building a digital twin platform, performing closed-loop verification on the maintenance priority sequence, and generating a unit maintenance scheme. The problem that multi-source data fusion of the wind turbine generator is difficult is solved, the health index quantification accuracy under the complex working condition is improved, the dynamic maintenance strategy is optimized, and closed-loop optimization of the maintenance scheme is achieved.

The invention relates to an AI server cooling water pump vibration quantity self-adaptive calibration test method and system, an intelligent test platform integrates a six-dimensional force sensor and a temperature compensation vibration exciter, the platform rigidity is automatically calibrated, a water pump-pipeline system transfer function is obtained, and a rotating speed-lift-modal frequency three-dimensional mathematical model is established; a three-axis MEMS accelerometer array is arranged at sensitive parts such as a water pump bearing seat and a motor shell, and vibration, current and pressure signals are synchronously collected; carrying out time-frequency domain signal processing by adopting variational mode decomposition in combination with self-adaptive S transformation, and extracting a 128-dimensional full-frequency domain feature vector containing a modulation side frequency band; effective values of vibration acceleration, speed and displacement are calculated through a frequency domain integration algorithm, and current harmonic interference is corrected through an electromagnetic vibration compensation model; and finally, a vibration health degree evaluation model containing 18 characteristic parameters is established based on a support vector machine, the test system comprises an intelligent test platform unit, a multi-source sensing unit, an edge calculation unit and a data management unit, and microsecond-level synchronous acquisition and GB-level data throughput are realized through a time sensitive network. And online incremental learning and automatic generation of a test report are supported. The method has the effect of improving the water pump vibration quantity test precision.

The invention relates to a power distribution network data intelligent analysis method based on data consanguinity and multi-modal fusion learning. The method comprises the following steps: S1, constructing a dynamically evolved data consanguinity topological graph; s2, designing a label-guided graph neural network architecture, embedding historical abnormal knowledge into a graph learning process, and outputting a deep semantic feature vector; s3, constructing a multi-modal fusion analysis framework, performing multi-dimensional feature fusion and data quality analysis, and identifying abnormal nodes; s4, designing a semi-supervised and incremental learning combined mixed training normal form, and performing model training and strategy optimization; and S5, based on the dynamic consanguinity topology constructed in the step S1 and the identified abnormal nodes, constructing a probabilistic reasoning framework, and fusing the model parameters obtained by optimization in the step S4 to realize quality abnormality root positioning and full-link visualization so as to form a complete data intelligent analysis scheme. According to the invention, efficient and accurate management of the topological data quality of the power distribution network is realized.

The invention discloses a communication network abnormal data intelligent diagnosis method and system based on a knowledge graph, and the method comprises the following steps: constructing a communication network knowledge graph: integrating multi-source heterogeneous data, including network topology data, equipment configuration data, historical fault cases and real-time monitoring data, and carrying out the entity extraction, relation modeling and attribute labeling, so as to obtain a communication network abnormal data database; constructing a knowledge graph containing network entities, entity relationships and attributes; and dynamically updating the knowledge graph: obtaining network operation data in real time through an incremental learning mechanism, updating an entity state, a relationship weight and an attribute value in the knowledge graph, and ensuring the real-time performance and the accuracy of the knowledge graph. A knowledge graph containing network entities, relations and attributes is constructed, multi-dimensional semantic modeling of network faults is achieved, and the problems of data isolation and relation fuzziness of a traditional method are solved.

The invention provides an equipment state deviation identification method based on self-supervision and incremental learning, and the method comprises the steps: S1, obtaining time sequence data of equipment in a fault-free state, and constructing a normal state model; s2, during operation, deviation detection is carried out on real-time data through the normal state model, and a deviation degree index is obtained; s3, comparing the deviation degree index with a preset threshold value, and judging an abnormal event; s4, determining new normal state data by manually verifying the abnormal event; and S5, updating the normal state model according to the new normal state data. The method does not need to depend on a fault sample, establishes an equipment normal behavior model through self-supervised learning, introduces a deviation index to quantify a state difference, and combines manual feedback and incremental learning to form a closed loop, so that the model has self-adaptability and long-term evolution ability.