496 results about "Module network" patented technology

A small-size vehicle detection deep learning model based on feature fusion of multi-scale modules is provided, which solves the problem of small-size vehicle image detection. The model includes a Backbone network, a Neck layer and a Head network, wherein a C2f_DCNv3 module based on the combination of deformable convolution v3 (DCNv3) and a cross stage feature fusion (C2f) module and an SPPF_LSKA module based on the combination of a spatial pyramid pooling fast (SPPF) layer and a large separable kernel attention (LSKA) module are introduced into the Backbone network; a C2f_SCConv module based on the combination of spatial and channel reconstruction convolution (SCConv) and a C2f module is introduced into the Neck layer; and a multi-scale kernel detection (MSK_Detect) module is introduced into the Head network.

The invention relates to the technical field of intelligent security operation and maintenance, in particular to an artificial intelligence network security system based on multi-modal large model training, which comprises a server fault diagnosis module, a network attack detection module, an endpoint security monitoring module, a key management optimization module and a threat analysis feedback module. According to the method, the fault prediction accuracy is improved through multi-dimensional data analysis, service interruption caused by sudden hardware faults is reduced, the network access frequency, source and instruction features are evaluated based on the server abnormality, the attack detection accuracy is improved, the misjudgment risk is reduced, the endpoint equipment execution behavior, resource calling and behavior sequence are extracted, and the service performance of the terminal equipment is improved. Fine-grained security monitoring is realized, attack traceability is enhanced, a key strategy is dynamically adjusted, security adaptability is improved, strategy lag risk is reduced, multi-level data is integrated to calculate threat behavior and attack fitting degree, threat assessment fineness and response speed are enhanced, and global security situation awareness is improved.

The invention relates to the technical field of power distribution cabinet detection, and discloses an intelligent detection system and method for a high-low voltage power distribution cabinet, and the system comprises a data collection module, a feature extraction module, a network construction and updating module, a fault positioning module, a positioning map generation module, and a self-healing control module. The method comprises the following steps: generating a multi-physical field coupling data set by synchronously collecting electromagnetic radiation spectrum and cabinet vibration waveform, performing time-frequency domain conjoint analysis to extract fault feature vectors, constructing and dynamically updating a fault propagation network, identifying key fault sources by combining betweenness centrality and historical case matching, and generating a fault positioning map; accurate self-healing control is realized by applying reverse electromagnetic pulse and adjusting contact pressure based on the atlas; according to the invention, the problems of low positioning precision and mechanical self-healing control means during detection of the high-low voltage power distribution cabinet are solved, and the fault detection efficiency and reliability are significantly improved.

The invention discloses an AI oT-based cloud network side-end federated learning system, and relates to the technical field of cloud network side-end federated learning, comprising the combination of Internet of Things, cloud computing and artificial intelligence technology, and through intelligent cooperative computing of cloud-side-end three-layer computing resources, the cloud network side-end federated learning is realized. The AI oT-based cloud network side-end federated learning system further comprises a cloud module, a network module, an edge module and a terminal module, and has the advantages that local model aggregation is firstly carried out on edge nodes, a part of calculation tasks can be completed at an edge end, a model result subjected to primary processing and aggregation is uploaded to the cloud, and the cloud network side-end federated learning is realized. Different edge nodes and terminal devices have different computing capabilities, the cloud end dynamically distributes training tasks to match tasks with different complexity degrees with devices with corresponding computing capabilities, and the edge nodes with strong computing capabilities can undertake more complex model training sub-tasks. And the terminal equipment with relatively weak computing power is responsible for some simple tasks such as data preprocessing and the like.

The invention relates to an LLM-Agent-based transformer substation TSN intelligent scheduling method and system. The system comprises an RAG-based TSN knowledge base management module, a TSN network global situation awareness module, an LLM-RAG-based TSN intelligent scheduling decision module, an LLM-based reflection module and a dynamic TSN scheduling simulation verification module. The RAG-based TSN knowledge base management module is used for constructing a knowledge base comprising a TSN scheduling algorithm, a TSN scheduling rule, a TSN protocol and a TSN network topology structure; the TSN global situation awareness module provides real-time data support for intelligent scheduling decision making; the LLM-RAG-based TSN intelligent scheduling decision-making module is used for carrying out intelligent scheduling decision-making by utilizing an LLM-RAG technology; the LLM-based reflection module ensures the rationality and effectiveness of a scheduling result; and the dynamic TSN scheduling simulation verification module performs simulation verification on the TSN scheduling result optimized by the LLM-based reflection module, so that intelligent scheduling of the transformer substation TSN network is realized, manual intervention is greatly reduced, and manpower cost investment is effectively reduced.

The invention discloses a network communication dynamic optimization method based on multi-module collaboration, and relates to the technical field of network communication, a dual-mode communication module is deployed at each network node, and the network communication dynamic optimization method comprises the following steps: each network node broadcasts own existence information and power line channel characteristics through an HPLC (High Performance Liquid Chromatography) channel of the dual-mode communication module; each network node scans surrounding wireless networks through an HRF channel of the dual-mode communication module and reports own wireless channel quality information to the gateway; the global state sensing module collects all information and constructs a global network view containing physical topology and a channel quality map. A dual-mode cooperation mechanism, a machine learning prediction model and a dynamic decision strategy can adapt to complex dynamic environments such as power line noise fluctuation and wireless interference change, and communication parameters can be autonomously optimized without manual intervention; and meanwhile, the modular design is convenient to expand to a multi-mode communication scene, and has a wide application prospect.

The invention discloses a network security threat detection system based on artificial intelligence. The system comprises a network security data acquisition module, a network security data optimization module, a network security threat prediction model establishment module, a model optimization module and a network security threat intelligent detection module. The invention relates to the technical field of network security data processing, in particular to a network security threat detection system based on artificial intelligence, which innovatively introduces weighted distance measurement driven by field correlation, adaptive neighbor number selection and a distance deviation threshold control strategy. A K-nearest neighbor method is improved to realize data completion, and the accuracy of a network security threat detection result is improved; a double-layer long-short-term memory network and self-adaptive threat threshold calculation are provided, and high-precision threat prediction of network security is realized; a sine smooth attenuation inertia weight strategy and a sine modulation contraction factor strategy are introduced to improve a particle swarm optimization algorithm, and the prediction accuracy and stability of the model are improved.

The invention discloses a fire detection and early warning method and system based on improved YOLOv8, and belongs to the technical field of image recognition, and the method comprises the steps: building a fire data set, marking the fire data set, and dividing the fire data set into a training set, a verification set and a test set; the YOLOv8 network is improved; the original YOLOv8 network and the improved YOLOv8 network are trained, evaluated, adjusted and tested, and an optimal model is determined; deploying the optimal model to an edge AI development board of a ground station; a real-time video stream of the unmanned aerial vehicle is obtained through the unmanned aerial vehicle camera and is transmitted to the edge AI development board for fire detection; a fire signal is detected by using the edge AI development board, and early warning is carried out through the fire alarm module; the system comprises a fire data set establishment and processing module, a YOLOv8 network improvement module, a YOLOv8 network training module, an optimal model deployment module, a fire detection module and a fire early warning module. The method has the technical advantages of high precision and light weight.

The invention relates to the field of industrial internet safety monitoring, and particularly discloses an industrial internet safety monitoring control system which comprises a network safety monitoring module, a control logic monitoring module, a traceability analysis module and an early warning execution module. The network security monitoring module analyzes a network attack risk level and an illegal access risk level through traffic baseline comparison, attack feature matching and risk model assessment; the control logic monitoring module identifies control logic abnormity based on the equipment operation state data and triggers a verification alarm; the traceability analysis module adopts a decision tree algorithm, combines the operation log, the behavior track and the version updating record, and judges that the tampering reason is misoperation, malicious tampering or system updating; and the early warning execution module triggers differentiated grading early warning according to the risk grade and tampering reason classification. The system can effectively identify external attacks, illegal access and logic tampering behaviors, and the security protection capability of the industrial internet is improved.

The invention provides a gas abnormal working condition monitoring and alarming system, relates to the field of gas monitoring and early warning, and adopts a monitoring system comprising a data preprocessing module, a spatial-temporal characteristic module, an ST-CGN network module, an edge intelligent optimization module and a result processing module. Wherein the data preprocessing module comprises a sliding window fragmentation method, a missing value interpolation method, a noise filtering method, a normalization method and a coding method; the spatial-temporal feature module extracts spatial correlation features, time dynamics features and causal propagation features, and fuses and splices the features into a three-dimensional feature tensor; the ST-CGN network module outputs a multi-working-condition probability vector corresponding to a sliding window fragment through the input of a three-dimensional feature tensor; and the edge intelligent optimization module realizes edge optimization of the ST-CGN network module. The system is used for monitoring and early warning gas abnormal working conditions, solves many problems existing in traditional gas pipe network monitoring, and has the advantages of being rapid in response, high in monitoring precision, high in cross-scene generalization ability and the like.

The invention discloses a network protocol analysis system based on a model context protocol MCP, and relates to the technical field of computer networks. The system comprises an MCP module, a network analysis module, a data source module, a context module, an MCP interface, a data source interface and a network analysis interface. An MCP module, a network analysis module, a data source module and a context module. All the modules work cooperatively through standardized interfaces to form a unified network protocol analysis platform. The system supports a context-driven tool calling process and a double-channel execution structure, control logic and data transmission are decoupled, and efficiency and expandability in a large-size Pcap data processing scene are improved. Through a unified semantic modeling and context state tracking mechanism, through analysis of a multi-layer protocol, multi-round traceable execution of a task and structured expression of an analysis result are realized, and infrastructure support is provided for a modularized, pluggable and reusable intelligent network protocol analysis platform.

The invention discloses an intelligent livestock necklace system with a self-adaptive adjustment function and a control method. The system comprises a behavior perception control module, a network perception transmission module, a group cooperation module, a strategy optimization module and an energy management module.

The invention relates to the technical field of communication scheduling command, in particular to a mining converged communication scheduling command system which comprises an information acquisition and indication module, an energy sensing communication control module, a network self-healing management module and a command scheduling integration module. According to the invention, the key information of the mining area is collected in real time and comprehensively processed, so that the accuracy and practicability of data are greatly improved, a mining area manager can quickly respond, the safety management efficiency of the mining area is improved, and the signal power and coding complexity of communication equipment are dynamically adjusted according to the environment and equipment energy state; the stability and efficiency of communication are obviously enhanced, rapid change of the internal environment of the mining area can be flexibly adapted, the self-healing network can rapidly reconstruct the network when finding a link quality problem, the continuity of communication is ensured, the fault recovery speed and reliability are improved, emergency response and resource configuration are optimized, and the network reliability is improved. The rescue efficiency of the mining area and the protection of life and property are enhanced.

The invention discloses a high-voltage SVG adaptive voltage control method based on deep learning, and belongs to the technical field of power equipment, and the method comprises the following steps: S1, multi-source data collection and preprocessing: the S1, multi-source data collection and preprocessing comprises a data collection module and a data preprocessing module, the data acquisition module comprises a high-precision sensor network and acquires various data in real time; the data preprocessing module comprises missing value filling through a sliding window method, wavelet transform denoising, multi-dimensional feature proof construction and normalization processing; s2, constructing a deep learning model, wherein the construction of the deep learning model in S2 comprises a model framework module, a network parameter module and a training strategy module. According to the method, the nonlinear characteristics of the power grid are captured through the deep learning model, the method can adapt to dynamic scenes such as power grid topological structure sudden change and new energy output intermittence in real time, and in the power grid with the new energy permeability exceeding 30%, the voltage regulation error can be reduced to + / -0.5% and is remarkably superior to + / -2% of a traditional method.

Certain aspects of the present disclosure provide techniques for a UE with multiple universal subscription identity modules (i.e., a multi-USIM device) or concurrent radio access technology (C-RAT) capabilities to receive and respond to a page from a network of a first USIIM of the UE while not activating a receiver on that network by receiving the page (from the network of the first USIM) on a receiver active on a network of a second USIM. A method that may be performed by a user equipment (UE) includes sending a request to register for paging events in a registration request message to an access and management function (AMF) of a network of a first universal subscriber identifier module (USIM) of the UE; sending a paging public user identifier (PPUID) of the UE, in a protocol data unit (PDU) session, to a paging server via a network of a second USIM of the UE, wherein the paging server is associated with the network of the first USIM; and receiving a message associated with the first USIM via the PDU session or another PDU session via the network of the second USIM.

The present application relates to the field of data processing, and disclosed are a network card device and a data transmission system, which are used for solving the problem whereby data transmission occupies the access bandwidth of a host memory. The network card device comprises a storage module; a descriptor queue address register; a calculation high-speed link hard core module, which is used for transmitting data sent by a host to a first data storage area, transmitting a descriptor sent by the host to a first descriptor storage area, and transmitting a trigger instruction sent by the host to a network data processing module; and the network data processing module, which is used for acquiring a base address from the descriptor queue address register when receiving the trigger instruction, reading a target descriptor in the first descriptor storage area based on the base address and the trigger instruction, reading target data corresponding to the target descriptor in the first data storage area, and outputting the target data to a network. The present application can reserve the access bandwidth of a host memory for other functions running on the host to use, so that the overall operation efficiency of the host is improved.

The invention discloses a digital power equipment management system which comprises an equipment automatic management module, a network topology visualization module, an intelligent alarm strategy engine, an automatic inspection execution module, a work order closed-loop management module, a multi-dimensional safety management and control module, an asset full-life-cycle data center and an AI fault prediction module. According to the invention, automatic discovery and full-life-cycle intelligent management of equipment can be realized, network topology dynamic visualization and fault rapid positioning capabilities are improved, a false alarm rate and operation and maintenance cost are reduced through triple verification intelligent alarm and adaptive routing inspection, a standardized operation and maintenance system is constructed in combination with a work order closed-loop process and multi-dimensional safety management and control, and the operation and maintenance efficiency is improved. And potential faults are warned in advance by using an AI prediction technology, so that the intelligent level and the operation reliability of power equipment management are comprehensively improved.

The invention provides an intelligent question answering system special for an electric power intranet, and relates to the technical field of electric power intranet security and protection. The intelligent question-answering system special for the electric power intranet comprises an account management module, a personnel information acquisition module, an identity verification module, a data security management module, a record management module, a network security monitoring module, an intrusion processing module, an intrusion operation misleading module and a central control unit. According to the intelligent question-answering system special for the electric power intranet, by comprehensively managing account and personnel information, enhancing identity verification, encrypting data security, monitoring the network in real time and the like, potential safety hazards are effectively solved, and the overall security of an electric power system is improved.

Disclosed in the present disclosure is a fast evaluation method of site seismic liquefaction hazard based on artificial intelligence algorithm: establishing a historical seismic and site information database, the database including a demand input module, a web crawler module, a data processing module, and a database module connected in sequence; a neural network model performs prediction to acquire a post-earthquake site dominant frequency; and, on the basis of the post-earthquake site dominant frequency, acquiring a site earthquake damage degree and seismic performance parameters. The present disclosure solves the problem of fast evaluating post-earthquake site earthquake damage and site seismic performance parameters, and can rapidly evaluate the site liquefaction or softening earthquake damage degree and site seismic performance parameters in given earthquake conditions.

The invention relates to the technical field of network security management, and discloses a federated learning-based dynamic encryption network security management system, which comprises a network data acquisition module, a federated learning processing module, a security state monitoring module and the like. The network data acquisition module acquires flow and encryption state data in real time, and generates a characteristic value and encryption strategy dynamic adjustment set; the federal learning processing module generates a personalized encryption strategy according to the traffic characteristic value; the safety state monitoring module screens abnormal safety data; the strategy matching module determines a target adjustment strategy through multi-dimensional matching; the strategy adjustment module drives the encryption engine to correct the strategy; and the strategy self-learning module updates the strategy correction factor based on the feedback data. The system realizes dynamic encryption strategy adjustment and self-optimization, improves the intelligence and self-adaptive capability of network security protection, and is suitable for distributed network security management.

The invention belongs to the field of network fault analysis, and particularly relates to a network fault intelligent prediction system based on dial testing, which comprises a dial testing module, a network testing module, a data acquisition and storage module, a signal prediction module, a fault analysis module and a monitoring and early warning module, according to the method, the principle of a graph neural network is creatively utilized, dynamic spatial-temporal feature extraction is performed on the dial test data, fusion features including overall network features and local features of network nodes are obtained, and rapid fault analysis is realized; according to the method, a bidirectional gating dynamic space-time signal prediction method based on a Pearson's correlation coefficient is creatively adopted, VMD decomposition is utilized, data proximity, periodicity and tendency characteristics are enhanced, Pearson characteristics are fused into a GAT model, a hidden space relation and potential characteristics between network nodes are highlighted, bidirectional gating characteristics are fused into a TCN model, and the prediction precision of the bidirectional gating dynamic space-time signal is improved. The problem of long-time dependence is solved, global features and dynamic changes are effectively extracted, and finally accurate processing of complex fault conditions is improved.

The invention discloses a multi-channel Ethernet and TTP data conversion hardware platform. The multi-channel Ethernet and TTP data conversion hardware platform comprises a core processing module, a network data interface module, a TTP data interface module, a configuration and storage module, a power module, a clock module and a monitoring module. According to the invention, a more reliable and more efficient data conversion and transmission hardware platform can be provided for industrial intelligent manufacturing and system-level power management. The core processing module adopts FPGA + ARM isomerism, so that efficient protocol data conversion, high-speed data analysis and calculation and interface type interaction functions can be realized, and high efficiency of operation is ensured; the network interface module and the TTP data interface module can realize multi-channel data transmission of a network layer and a physical layer, so that the reliability of the data is ensured; and the configuration and storage module can realize storage of configuration information and monitoring acquisition data, so that the effectiveness of the information is protected. Compared with a traditional data conversion platform, the data conversion platform has the advantages of reliable protocol conversion, high function expansibility, large storage space, low data processing delay and the like.

The invention discloses a fiber yarn variety identification method and device, relates to the technical field of cheese variety identification, and solves the problem that in the prior art, due to the fact that an identification model is complex in structure, high in computing resource requirement and large in storage and memory occupation, deployment cannot be conducted on edge equipment. The method comprises the following steps: constructing a YOLO-FSL network model, training and compressing the YOLO-FSL network model by using a data set to obtain a lightweight variety identification model, identifying yarn varieties by using the lightweight variety identification model, and identifying yarn varieties by using a lightweight feature extraction module C3k2-FG and an HSFPN network as a neck network and an SRD detection head based on a weight sharing mechanism. A lightweight YOLO-FSL network model is constructed, a lightweight variety identification model capable of identifying the cheese variety is obtained after training, compression and fine adjustment, the parameter quantity, the calculation quantity and the model size are remarkably reduced while the high-precision model is kept, and then edge equipment deployment can be achieved.

The invention discloses a network equipment dynamic data authority grading authentication management platform, and relates to the field of data authority. The platform comprises a network equipment position area sensing module, a network equipment multi-modal data analysis module, a network equipment compliance security authentication module, a multi-authority historical access behavior analysis module, a network equipment dynamic authority affirmation marking module and an access control execution module. The method comprises the following steps: performing multi-dimensional position secret-related grading according to a physical position, analyzing data content of network equipment in real time, identifying secret-related data flow of the network equipment in real time, performing multi-dimensional data secret-related grading, and performing compliance security authentication on the network equipment at the same time. According to the method, the confidential level of the network equipment is evaluated by combining a feedback mechanism of historical access behavior analysis of multi-authority personnel, a corresponding authority strategy is bound based on a position, data and equipment security ternary framework, and dynamic authority grading authentication is intelligently performed on the network equipment.

A container simulation network system based on a link autonomous agent is composed of a visual interaction module, a control plane module, a network simulation module and a resource management module. The modules perform information interaction and task collaboration through a uniform interface so as to realize automatic construction and management of the simulation network, a two-layer architecture design of network bridge containers and user containers is adopted, and each user container corresponds to one network bridge container; a link autonomous agent module is deployed in the network bridge container and is used for collecting a link state, calculating a performance adaptation degree and executing a scheduling operation bit; the visual interaction module realizes visual monitoring and interaction control of a simulation state; the control plane module is responsible for simulation scene analysis, link parameter generation and global scheduling management; and the network simulation module is responsible for calling a container generation tool to generate a specific container instance and carrying out resource monitoring and adjustment on the container.

The invention relates to the technical field of new energy automobile battery management, in particular to a wireless BMS management system with high anti-interference capability, which comprises a master node, a plurality of slave nodes, a dual-band communication module, a Mesh network topology structure, a broadband ceramic antenna array and an AI channel prediction module. The wireless BMS management system has remarkable advantages in the aspects of anti-interference capability, communication stability, network flexibility and the like, can effectively meet the requirements of modern battery management, and provides powerful guarantee for safe and efficient operation of an electric vehicle and an energy storage system.

The invention discloses an intelligent fire safety monitoring system which comprises a sensing layer, a network layer, a platform layer and an application layer. The sensing layer comprises a sensor module and a camera module; the network layer comprises a data transmission module; the platform layer comprises a data processing module, a data storage module and a threshold adjustment module; the application layer comprises a monitoring interface module, an alarm system module, an emergency response module, a remote monitoring and commanding module and a man-machine interaction module; the data processing module receives data from the data transmission module of the network layer and transmits the processed data to the data storage module and the application layer, and the threshold adjustment module receives feedback data from the application layer; according to the invention, the self-adaptive algorithm is introduced, and the self-adaptive algorithm unit can learn rules in historical data and automatically adjust the fire early warning threshold according to the current environmental parameters and the change of the time period, so that the accuracy and reliability of fire monitoring are improved, and the change condition of the fire risk is reflected more accurately.

The invention discloses an aviation radiation dose global distribution real-time calculation system based on ground detection equipment, which belongs to the technical field of radiation detection and comprises a ground radiation detection module, a dose simulation module and a data correction module. The ground radiation detection module is used for continuously collecting ionizing radiation data of an atmospheric bottom layer in a calm state and a solar high-energy particle event so as to improve data precision and a signal-to-noise ratio under a weak dose, and the dose simulation module receives latitude and longitude, height and time parameters of an aviation flight path, and calculates a standard atmospheric model and a cosmic ray energy spectrum in combination with the standard atmospheric model and the cosmic ray energy spectrum. The data processing module is used for calculating theoretical dose rates and particle type distribution of different height levels based on a Monte Carlo method and outputting the theoretical dose rates and the particle type distribution, and the data correction module is used for dynamically correcting parameters and weights of simulation results by combining measured data based on a deep learning algorithm of deep learning, so that adaptive fusion of the simulation data and the measured data is realized; and the ground radiation detection module comprises a semiconductor radiation detector, a data acquisition and preprocessing unit, a remote control module and a network communication module. The system has the characteristics of high real-time performance and wide coverage range.

The invention provides a description generation method based on continuous zero samples. The description generation method comprises the following steps: sequencing plain text training corpora according to the difficulty of continuous learning; setting that each training step t only can access the t-th corpus, and for the t-th learning step, generating a synthetic image for an input text of the t-th corpus by using a diffusion model; extracting image embedding of the synthesized image by using a CLIP image encoder; retrieving a group of text description embedding with similar semantics in a text corpus by using image embedding; image embedding and text description embedding are input into a fusion module network, and the output of a fusion module is defined as a soft prompt of a large language model; for an input text, key entities are extracted from the text through word frequency statistics so as to construct a soft prompt and a hard prompt of the large language model; hard prompts and soft prompts are input into a large language model network, and key parts in input are concerned through an attention mechanism.

The invention relates to the technical field of electroplating wastewater treatment, and discloses an electroplating wastewater pollution early warning and regulation system based on Internet of Things, comprising: a sensing layer for collecting characteristic parameters of an electroplating wastewater treatment link through distributed sensing nodes, the characteristic parameters including a water quality parameter, an equipment operation parameter and an environmental parameter, a wireless communication module is arranged in the sensing node; the network layer is used for constructing a wireless communication network and transmitting the parameters acquired by the sensing layer to the application layer through a multi-hop relay; and the application layer receives and processes the parameters transmitted by the network layer to realize pollution early warning, pollution risk prediction and regulation and control instruction generation, and the regulation and control instruction is fed back to the execution layer through the network layer. By adopting a channel frequency hopping mechanism, when it is monitored that the interference intensity of a current communication channel exceeds a preset interference threshold value, the standby channel can be automatically switched, instruction transmission of an application layer and an execution layer is not interrupted, and it is guaranteed that data such as water quality parameters and equipment operation parameters collected by a sensing layer are continuously and stably transmitted to the application layer.