2448 results about "Network architecture" patented technology

The invention relates to the technical field of cable operation state detection, and particularly discloses a multi-source data driven cable operation state comprehensive evaluation method, which comprises the following steps of S1, adopting a layered distributed sensing network architecture, and deploying three types of core sensors at key nodes of a cable, through space-time calibration of the multi-source heterogeneous sensor, data consistency is improved, fusion deviation is eliminated, the problem of data islands of a traditional system is solved, and a precise evaluation foundation is laid; noise suppression and dynamic correlation modeling are adopted, environmental interference is stripped, a vibration and displacement coupling relation is quantified, limitation of a single parameter is broken through, heterogeneous fault features are captured, and evaluation comprehensiveness and sensitivity are improved; a self-adaptive threshold mechanism is constructed based on environment weight and historical data, the bottleneck of a fixed threshold is broken through, an evaluation standard is corrected along with equipment aging and environment change, misjudgment is avoided, and diagnosis robustness in different scenes is enhanced.

The invention discloses a trajectory optimization method based on a multi-source heterogeneous positioning data fusion algorithm, and relates to the technical field of intelligent navigation and high-precision positioning, multi-modal data are acquired through a multi-modal sensor array, positioning redundancy of scenes such as tunnels and indoor scenes is enhanced, a weight distribution strategy is dynamically adjusted through an Actor-Critic network architecture, and the positioning accuracy is improved. The state space input comprises an environment semantic tag, a historical error sequence and a real-time noise variance, the output action space is continuous weight distribution of each data source, a multi-target reward function optimization strategy is combined, scene adaptability is realized, a local SLAM map, inertial navigation error parameters and a weight distribution strategy are shared in real time based on a V2X protocol, and the real-time performance of the system is improved. According to the method, a single device accumulative error is compensated by using adjacent vehicle data, a terminal locally trains an error compensation model, parameters are uploaded to a cloud end through differential privacy encryption, the cloud end adopts a FedAvg algorithm to aggregate a global model and issue the global model, the error difference between devices is inhibited, and dynamic road network updating and scene differentiation model distribution are supported at the same time.

The invention discloses a mechanical equipment state monitoring method and system based on multiple sensors, and the method comprises the five core steps: multi-modal data collection and preprocessing, dynamic feature fusion, adaptive threshold diagnosis, digital twin fault tracing and predictive maintenance decision. All-domain coverage of equipment is realized through a three-layer sensor network architecture, the problems of data synchronization and interference resistance are solved by utilizing a temperature and vibration integrated sensor, deep fusion and anomaly detection of multi-source data are realized in combination with an attention mechanism, a Gaussian mixture model, a three-dimensional convolutional neural network and the like, and finally a precise maintenance strategy is generated through digital twinning and reinforcement learning. The multi-sensor-based mechanical equipment state monitoring system comprises a sensor network layer, an edge computing layer, a cloud platform layer and a man-machine interaction layer, supports federated learning to protect data privacy, improves real-time diagnosis capability through edge-cloud collaboration, and enhances a reality interface to realize intelligent operation and maintenance interaction.

The embodiment of the invention provides an end-to-end automatic driving control method and device based on multi-camera fusion, and multi-view target detection and tracking are realized through spatial transformation and coordinate mapping by combining front wide-angle camera information and left and right wide-angle camera information. A multi-view feature fusion network architecture is designed, the multi-view feature fusion network architecture comprises three sub-networks of feature extraction, dynamic weight distribution and feature fusion, and the fusion weight is dynamically adjusted based on image definition, detection confidence and view overlapping degree. A geometric consistency constraint between visual angles and a reconstruction loss function are introduced, a deep neural network model is constructed, abnormal conditions such as camera shielding are effectively handled, and an accurate control instruction is output. According to the method, the defects of the traditional technology in the aspects of multi-view information fusion, shielding processing and the like are overcome, and the sensing ability and the control reliability of the automatic driving system are remarkably improved.

The invention discloses a multi-label electrocardiogram classification method based on self-supervised pre-training and multi-modal semantic alignment, which belongs to the technical field of artificial intelligence, and comprises the following steps: realizing self-supervised pre-training of unlabeled data through a single-modal contrast enhancement network, generating global and local contrast views by adopting a multi-scale random cutting strategy, and classifying the global and local contrast views in a multi-scale random cutting mode; in combination with a teacher-student network architecture, the potential invariance features of the ECG signals are learned while negative sample dependence is avoided, the problem of annotation data scarcity is effectively relieved, and the feature robustness is improved. A multi-modal fusion mechanism based on label semantic guidance is provided, a time domain signal and a frequency domain time-frequency graph are mapped to a unified semantic space through fine-grained semantic alignment, local feature enhancement and cross-modal complementary information fusion are realized by using a cross attention mechanism, and the problem of semantic difference caused by modal heterogeneity in a traditional method is overcome. A multi-label comparison loss function based on a disease co-occurrence relation is proposed, a category discrimination boundary is dynamically optimized by modeling a label co-occurrence probability, the feature separability of a tail category is improved while the head category discrimination ability is enhanced, and the problem of sample category imbalance in a multi-label scene is remarkably relieved.

Generally, the present disclosure is directed to efficient neural network architectures, techniques for constructing new efficient neural networks, and approaches to low-latency execution of neural networks. In an example aspect, the present disclosure provides a more powerful configuration of an inverted bottleneck block that maintains an efficient execution profile. An example search system can parameterize a universal inverted bottleneck block with a first parameter that categorically activates a spatial mixing operation in the unexpanded state. In this manner, for instance, a large variety of different network architectures can be explored using a relatively compact search space that admits high levels of parameter sharing. Further, the present disclosure introduces an efficiency-optimized multi-query attention block.

The invention provides a three-dimensional seismic fault identification method based on double-attention multi-scale fusion U-Net. The method comprises the following specific steps: constructing a U-shaped network architecture comprising an encoder, a decoder and jump connection; a constructed multi-scale feature fusion module is embedded in the first level of the encoder, and the extraction capability of fault features of different scales is enhanced through a multi-branch structure; introducing the constructed hole fusion modules into the second and third levels of the encoder, designing and expanding a receptive field by using multiple expansion rates, and capturing fault structures of different scales; a double-attention parallel mechanism is integrated in jump connection, and the sensitivity of channel attention and space attention to fault features is improved; constructing a combined loss function; and finally, performing three-dimensional seismic data training and reasoning based on the optimized model to realize high-precision fault identification. The method has high generalization and accuracy, and especially has good performance in the aspect of seismic image fault identification containing a large fault scale span.

The invention discloses a heterogeneous computing multi-target adaptive task scheduling method based on deep reinforcement learning, and the method comprises the following steps: S1, constructing a multi-dimensional dynamic perception model of a heterogeneous computing environment, and collecting and computing node performance indexes, task feature parameters and network states in real time; s2, defining a reward function as a multi-target weighted combination, fusing task completion time, energy consumption, resource utilization rate and cost, and dynamically adjusting the weight by a fuzzy comprehensive evaluation algorithm; s3, establishing a dual-channel deep reinforcement learning network architecture based on an attention mechanism; s4, establishing an adaptive exploration mechanism, combining an epsilon-greedy strategy and entropy regularization, and balancing exploration and utilization; the method has the beneficial effects that dynamic balance of multiple indexes such as task completion time, energy consumption and resource utilization rate is realized through combination of deep reinforcement learning and multi-objective optimization, a dual-channel network and a cross attention mechanism are adopted, and a task time sequence characteristic and a topological dependency relationship are modeled at the same time, so that a scheduling strategy is more accurate.

The present invention relates to the technical field of railway inspection, and relates in particular to a railway foreign-object intrusion detection method and system based on deep learning. The method comprises: S10, acquiring image data to undergo detection; and S20, inputting the image data into a trained attention semantic segmentation network to obtain a foreign-object detection result. The attention semantic segmentation network is obtained by first training a preset attention semantic segmentation network architecture using a preconfigured data set, and then re-training the attention semantic segmentation network on specified image data using an adaptive correction algorithm. A backbone network architecture is obtained by inserting a specified attention mechanism at a specified position within a pre-selected residual neural network, and using three parallel dilated convolutions as an initial convolutional layer in the residual neural network. A dual-branch decoder combines an edge recognition branch and a semantic segmentation branch. The method is applicable to various scenarios, achieves improved detection accuracy and maintains a lightweight design.

The invention discloses an industrial internet of things-oriented equipment digital twin operation and maintenance management system, and relates to the technical field of equipment management. The system comprises a data acquisition and preprocessing module, a digital twin model construction module, a data transmission and storage module, a state monitoring and fault diagnosis module, an operation and maintenance decision and optimization module and a visual interaction module. The data acquisition module adaptively acquires data through a sensor and preprocesses the data; the model construction adopts multi-scale and multi-model fusion; a hybrid network architecture and an encryption technology are used for transmission and storage; performing feature fusion and transfer learning for monitoring diagnosis; reinforcement learning and multi-agent collaboration are used for decision optimization; visual interaction supports VR / AR fusion. According to the invention, multiple modules work cooperatively, accurate acquisition, efficient transmission and storage of data are guaranteed, and accurate fault diagnosis and scientific operation and maintenance decision are realized through an advanced algorithm; the operation experience is improved through visual interaction; the method also has energy consumption optimization and supply chain cooperation capabilities, and can improve the operation and maintenance efficiency of industrial equipment and enterprise benefits.

The invention discloses a multi-unmanned aerial vehicle negotiation anti-collision method and system based on 5G unmanned aerial vehicle communication, and belongs to the technical field of multi-unmanned aerial vehicle risk avoiding. The method comprises the following steps: collecting multi-source flight state data; after the multi-source flight state data are fused through Kalman filtering, an airspace situation map is constructed; a 5G network architecture is constructed, and real-time data interaction between the unmanned aerial vehicles is realized; a dynamic negotiation protocol is constructed, and fast exchange of obstacle avoidance proposals when collision risks are triggered by multiple machines is supported; and constructing a distributed collaborative obstacle avoidance algorithm, and generating a smooth obstacle avoidance path in real time. According to the invention, through a dynamic negotiation protocol, multiple machines are supported to rapidly exchange obstacle avoidance proposals when a collision risk is triggered; a global optimal strategy is generated through edge node intelligent arbitration; the priority mechanism gives consideration to task urgency and environment dynamic change, ensures that a high-value task unmanned aerial vehicle passes preferentially, balances fairness through distributed voting, avoids path stiffness caused by decision conflict or resource competition, and realizes efficient and reasonable collaborative decision.

The invention relates to the technical field of intelligent weather forecast, and discloses a WRF wind speed simulation correction method, system and device and a storage medium, and the method comprises the steps: obtaining multi-source basic meteorological data, and carrying out the preprocessing; constructing a fan field feature matrix according to the fan field distribution data; inputting the preprocessed data and the wind field feature matrix into a multi-source space-time convolution fusion module, and extracting and fusing time sequence features and space correlation features; inputting the fused features into a normalized convolutional neural network introducing physical constraints, and extracting spatial-temporal features of the wind speed field; and performing channel splicing on the spatial-temporal characteristics of the wind speed field and to-be-corrected wind speed data, inputting the spliced data into a multi-scale convolutional wind speed correction network, and obtaining a correction result of the wind speed simulation data through encoding and decoding operations. Through a multi-source data dynamic fusion mechanism, a spatial-temporal feature collaborative optimization algorithm and a multi-scale correction network architecture, system errors of traditional numerical mode simulation are remarkably reduced, and the accuracy of wind speed forecasting is greatly improved.

The invention discloses a cross-regional vehicle-mounted network switching method and system based on digital twinning, and a medium. The method comprises the following steps: constructing a cloud side end digital twinning network architecture; acquiring vehicle state data and traffic environment sensing data through the terminal sensing layer; carrying out data aggregation on the vehicle state data and the traffic environment perception data through edge computing nodes to obtain a local digital twinborn model; meteorological data, operator base station data and a local digital twinborn model are integrated through a cloud digital twinborn center to obtain a global digital twinborn model, and the traffic congestion condition and the network congestion condition of each road network region in a future time period are predicted according to the global digital twinborn model. Determining a network switching strategy according to the traffic congestion condition and the network congestion condition; and performing a network switching decision on the target vehicle through the edge computing node. According to the method, the timeliness and accuracy of cross-regional vehicle-mounted network switching are improved, so that the stability and reliability of the vehicle-mounted network are improved, and the method can be applied to the technical field of the Internet of Vehicles.

The invention relates to the technical field of security management, in particular to an AI-based intelligent security remote inspection and control method, which comprises the steps of front-end intelligent equipment deployment, intelligent inspection execution, data processing and intelligent analysis and remote control and linkage response. Compared with the defects of incomplete information and high false alarm rate due to the fact that equipment appearance anomaly detection mainly depends on single-modal data in the prior art, the scheme adopts multi-modal data preprocessing and feature extraction, constructs a cross-modal alignment network architecture, dynamically fuses laser radar, RGB-D images and IMU data by using a self-attention mechanism, and improves the detection accuracy of the equipment appearance anomaly. According to the method, more accurate equipment appearance abnormity identification is realized by combining environment self-adaptive adjustment, and the model is compressed and deployed to an edge end through a knowledge distillation technology, so that real-time reasoning and analysis are realized, the intelligent level of security inspection and the accuracy of abnormity detection are remarkably improved, the false alarm rate is reduced, and the practicability and reliability of the system are enhanced.

The embodiment of the invention provides an automatic driving method and device based on a multi-dimensional reward function, and the method and device achieve the control of a driving motion through the combination of an imitation learning framework and a reinforcement learning framework, collection of environment information through a plurality of cameras, and construction of a strategy generation network and a discriminator network. A multi-dimensional reward function model is designed, behaviors such as red light running, line pressing, lane departure and collision are detected and evaluated in real time, and a driving behavior reward and punishment matrix is constructed. Based on an actor evaluation network architecture, environment information and a navigation instruction are input into an actor network to generate an optimal driving action, and the action value is evaluated through the evaluation network to realize dynamic parameter optimization. According to the method, the defects of the traditional technology in the aspects of driving behavior evaluation, action value judgment and the like are effectively overcome, and the safety and reliability of the automatic driving system are remarkably improved.

An unmanned aerial vehicle (UAV) agent for forming a two-tier hybrid smart agriculture network architecture is provided. The UAV agent is configured to support two communication protocols, a short-range communication protocol and a long-range communication protocol. The agriculture sensor gathers agriculture data such as soil temperature and participates in a first-tier short range wireless communication network to send the collected data to at least one UAV agent. A UAV agent participates in one or more first-tier short-range communication networks to pick up sensor data from agriculture sensors in one or more clusters and also participates in a second-tier long-range communication network to route the collected sensor data to at least one cloud server. The tasks in both tier networks are formulated as optimization problems to achieve optimal data delivery and solved by using multi-agent reinforcement learning (MARL), which is implemented by the invented Focus Coordination Multi-Agent Deep Deterministic Policy Gradient (FC-MADDPG) algorithm.

The invention relates to a microphone array sound source positioning method and system based on cross-correlation-beam forming closed-loop optimization, and belongs to the technical field of sound source positioning. The method comprises the following steps: collecting multichannel sound signals through a microphone array and preprocessing the multichannel sound signals to extract time-frequency features and suppress noise interference; time delay information among the microphones is estimated by adopting a generalized cross-correlation phase transformation algorithm, and an optimization strategy is introduced to improve estimation stability and anti-interference performance; enhancing the target sound source signal in combination with a minimum variance undistorted response beam forming algorithm and an adaptive Kalman filtering mechanism; constructing a closed-loop feedback optimization mechanism based on the beam output signal to realize feedback adjustment; and adopting a hybrid network architecture, taking the beam output signal amplitude spectrum as input, and outputting the frequency spectrum or mask of the obtained target sound source signal. The method has the advantages of high calculation efficiency, high positioning precision and strong anti-interference capability, and is suitable for real-time acoustic signal processing in a complex environment.

The invention provides a motor defect identification method fusing time sequence space feature extraction and reinforcement learning. The method comprises the following steps: building a Transform-GAT model architecture T-GAT, and learning time sequence relevance and spatial topological structure features by the T-GAT to form a space-time composite representation vector; designing a dual-network architecture reinforcement learning weighted fusion mechanism of a strategy network and a value network, and after the strategy network and the value network of a parallel structure receive the composite vector, constructing a strategy function to select a defect type with the maximum probability; designing potential energy function quantization parameters, and constructing a reward function to generate a reward in combination with a difference value; taking a reward function as a target, training and optimizing model parameters of the dual-network architecture in an off-line manner, running a real-time decision in an on-line manner, and storing a tetrad to an experience pool to form a'perception-decision-feedback-update 'closed-loop mechanism; according to the method, motor defect identification is realized, downtime is reduced, and motor operation reliability and equipment operation efficiency are improved.

The invention relates to the field of copyright authentication of non-perpetual culture elements, and discloses a block chain copyright authentication method for non-perpetual culture elements, which comprises the following steps: S1, performing multi-dimensional data acquisition through a distributed storage system, performing structured data acquisition and feature coding on the non-perpetual culture elements, and storing the structured data in a database; comprising the following steps: multi-modal data fusion acquisition, dynamic metadata standardization processing, and generation of unique digital identifiers and cultural feature codes; and S2, constructing a hybrid block chain architecture, constructing a hierarchical network architecture comprising a permission chain layer and an open chain layer, and realizing heterogeneous chain data synchronization through a cross-chain interoperation protocol. Through multi-modal data acquisition and dynamic metadata structured processing, the spatial form, the process time sequence and the oral context of non-residual elements are completely recorded, digital twin bodies with microscopic details and semantic association are formed, the problem of fragmentation of traditional archiving is solved, and a unique block chain identifier and two-way index mapping are combined, so that the method is simple and convenient to implement. And the originality of the cultural heritage in virtual restoration and cross-domain propagation is ensured.

The invention relates to the technical field of industrial distributed control system architecture, and discloses a dual-redundancy design method for a virtual controller VDPU, which comprises the following steps of: deployment of a four-network-port redundant network architecture: dividing four physical network ports into two groups of independent channels, and simultaneously configuring a network link automatic switching mechanism, the working states of the four network ports are monitored in real time, then double isolation of a physical layer and a protocol layer is achieved, and an independent heartbeat transmission channel is established; main / standby heartbeat monitoring and switching control: the main / standby controller periodically interacts state information packets through a special heartbeat channel, and establishes a three-level fault determination mechanism at the same time to perform seamless main / standby switching; a multi-mode data synchronization mechanism is implemented, namely initialization synchronization, increment synchronization during operation and exception recovery synchronization are started, separation of heartbeat signals and data synchronization can be realized through two groups of independent channels, and the fault-tolerant capability and the switching efficiency of the system are improved, so that the operation stability of the system is guaranteed.

The invention relates to the technical field of power distribution station supervision, in particular to an intelligent auxiliary and artificial intelligence visual gateway control method and system for a station building, which monitors network state parameters in real time, calculates a network quality score and dynamically selects an optimal transmission path based on a reinforcement learning algorithm. Comprising a high bandwidth mode, a low delay mode and a disaster recovery backup mode, stage processing and dynamic compression are performed on transmission data according to data types and network quality scores, link switching and a local caching mechanism are automatically triggered when a network is abnormal, and finally, multi-gateway load balancing and visual monitoring are realized through a software defined network architecture. The problem of transmission delay or interruption caused by the fact that a static data transmission framework cannot adapt to network state changes in the prior art is effectively solved, and the reliability and the real-time performance of data transmission are remarkably improved through dynamic path optimization, intelligent hierarchical transmission and a fault self-healing mechanism.

The invention relates to the field of scene rendering, and particularly discloses a three-dimensional scene optimization method and system for collaborative rendering of dynamic LOD and view cone rejection based on space-time prediction, and the method comprises the steps: dynamically calculating the visibility frequency of an object, and dynamically adjusting the loading and rendering modes of objects with different priorities; predicting a view cone range of multiple frames in the future by using an LSTM network architecture; dividing the scene into uniform grid blocks, and then performing grading elimination; optimizing a heterogeneous computing pipeline; rendering the execution process; the system comprises a dynamic LOD and visual cone rejection collaborative optimization module, an LSTM visual cone prediction module, a block-level mixed rejection module, a heterogeneous calculation pipeline module, an optimization CPU-GPU task allocation and data transmission module and a rendering execution control module. According to the method, the technologies of collaborative optimization of dynamic LOD and view cone removal, block-level mixed removal, heterogeneous calculation assembly line optimization and the like are adopted, so that unnecessary rendering calculation and resource loading are reduced, and the rendering efficiency is improved.

The invention belongs to the technical field of all-dielectric metamaterial optical devices and machine learning, and discloses an all-dielectric metasurface target spectral response reverse design method based on deep learning. And realizing efficient prediction of the transmission spectrum by using a convolutional neural network-recurrent neural network-residual network architecture. A fitness function is designed, and a machinable structure is generated by aiming at single-peak and multi-peak target wavelength optimization and combining a linear and shape optimization strategy. The method breaks through the limitation of spectrum dependence and fixed structure type of the traditional reverse design, realizes on-demand design, and remarkably improves the design efficiency and processing compatibility of the integrated photonic device.

The invention relates to the technical field of Internet security, in particular to a dynamic trust evaluation method and system for heterogeneous fusion network nodes, and the method comprises the steps: carrying out the dynamic modeling of a network into a space-time heterogeneous graph based on a network architecture of a cloud-edge cooperative industrial Internet of Things, and enabling the space-time heterogeneous graph to comprise a sensing layer node, an edge layer node and a cloud node; performing joint embedding characterization on the dynamic topology by using a knowledge graph embedding technology, and extracting node attributes and spatial-temporal characteristics of interaction; multi-source feature interaction is fused based on a cross attention mechanism, and the weight of a trust factor is dynamically adjusted through a multi-time-slot feature; and updating the trust evaluation model through an incremental learning strategy, and establishing a dynamic mapping relationship between a trust evaluation value and an equipment time sequence behavior mode. According to the method, continuous and accurate evaluation of the dynamic trust of the network nodes under the zero-trust architecture is realized by fusing the space-time diagram representation learning and the cross attention mechanism, and the anti-attack capability and the safety performance of the industrial Internet of Things are enhanced.

The invention provides a hyperspectral snapshot compression imaging method and system based on a space-spectrum prior decoupling model, high-quality reconstruction is realized through decoupling optimization and a deep expansion network, and the method comprises the following steps: constructing a training data set containing a compression measurement image and a corresponding reconstruction spectrum; establishing an objective function fusing space and spectrum prior, converting the objective function into constrained optimization, and converting the constrained optimization into three sub-problems of linear reconstruction, space prior and spectrum prior by adopting a semi-quadratic splitting method; a deep expansion network is designed to alternately solve sub-problems: a linear sub-problem is solved through analysis, and a space / spectrum sub-problem is subjected to implicit prior modeling through a private network, so that end-to-end reconstruction is realized; a mixed loss function is adopted to optimize model parameters, and images can be reconstructed in real time after training is completed. Space and spectrum prior decoupling is carried out, space structure details and spectrum features are respectively captured through an independent network architecture, the problem of mutual interference of joint modeling in a traditional method is solved, and high-quality spectrum image reconstruction is realized.

The invention discloses a construction method of a self-adaptive gated spectrum-space-graph collaborative fusion network. The construction method comprises the following steps: step 1, constructing a network overall architecture; step 2, spatial branching-hierarchical spatial feature modeling is carried out; step 3, spectrum branching-adaptive spectrum feature optimization; step 4, designing an adaptive gating fusion module AGFM; the invention provides a spectrum-space-graph collaborative fusion network, which is a double-flow architecture, and solves the problems of complex space-spectrum interactive modeling, spectrum redundancy and low calculation efficiency in hyperspectral image classification. According to the network, through integration of hierarchical spatial feature learning, adaptive spectrum optimization and a dynamic cross-modal fusion mechanism, complementary advantages of a graph attention mechanism, intelligent agent self-attention and data-driven spectrum modeling are effectively combined; experimental results on three reference data sets verify the advanced performance of the SGCFN, and ablation studies prove that each module has an irreplaceable effect on enhancing classification robustness.

The invention relates to the technical field of geophysical exploration, in particular to a multichannel deep learning magnetotelluric inversion method based on physical information constraint. The method comprises the following steps: generating a synthetic data set containing a geoelectric model and forward modeling response thereof, and adding a noise simulation actual observation condition; constructing a hybrid network architecture combining Transform and U-Net, taking apparent resistivity and impedance phase as dual-channel input, extracting global features by using an encoder, gradually recovering spatial resolution through a decoder, and outputting an underground resistivity model; network training adopts a composite loss function fusing model loss and data loss, and an inversion process is constrained by introducing a magnetotelluric forward modeling physical rule, so that a result is ensured to fit observation data and conform to a physical mechanism; after training is completed, preprocessed actual measurement data are input into the model, and a resistivity image can be directly obtained. The method is used for geological structure identification and reservoir interpretation, and the inversion precision and reliability are effectively improved.

The invention relates to the technical field of network security, and discloses a network traffic anomaly detection method and system based on a knowledge graph, and the network traffic anomaly detection method comprises the following steps: protocol perception metadata feature extraction: extracting metadata features which do not involve content privacy from encrypted network traffic through a deep packet inspection technology, comprising flow statistical characteristics, time sequence characteristics and connection relation characteristics; multi-level knowledge graph construction: based on the extracted metadata features, according to a network architecture, respectively constructing corresponding knowledge graphs on a device layer, a gateway layer and a cloud layer, and respectively representing device behaviors, network activities and global security information; the method does not depend on flow decryption operation, effective recognition of abnormal behaviors in the encrypted flow is achieved only by analyzing the metadata features of the network flow, and therefore the detection accuracy is improved.

Systems, methods, and computer readable media related to: training an encoder model that can be utilized to determine semantic similarity of a natural language textual string to each of one or more additional natural language textual strings (directly and / or indirectly); and / or using a trained encoder model to determine one or more responsive actions to perform in response to a natural language query. The encoder model is a machine learning model, such as a neural network model. In some implementations of training the encoder model, the encoder model is trained as part of a larger network architecture trained based on one or more tasks that are distinct from a “semantic textual similarity” task for which the encoder model can be used.

The invention discloses a wind power prediction system for optimizing a neural network based on a genetic algorithm, relates to the technical field of new energy power system prediction, and improves the precision and adaptability of wind power prediction by fusing the genetic algorithm and a deep neural network. The system adopts multi-objective genetic optimization, randomly initializes a neural network parameter combination, evaluates the fitness by taking a prediction error and model complexity as double objectives, and screens out an optimal network architecture through evolution operation; in the aspect of neural network training, the system adopts an LSTM and TCN hybrid network as a basic model, dynamic weighting input features of a meteorological attention mechanism are combined, a learning rate and regularization parameters are optimized by using a genetic algorithm, model convergence is accelerated, and overfitting is prevented; in addition, for the space-time imbalance of the wind power data, a generative adversarial network is introduced to generate synthetic data in an extreme weather scene, and the generalization ability of the model is enhanced.