193 results about "Node level" patented technology

The invention discloses an intelligent question and answer reply generation method and device based on a knowledge graph. The method comprises the steps of obtaining knowledge graph data and performing structured processing to construct a node level table and a multi-modal index table; reconstructing a target problem according to the historical session; calling a large language model through a thinking chain to generate a reply, and embedding a structured fact to inhibit illusion; and an active wake-up mechanism is configured to improve interactivity. According to the method, through multi-modal data mapping and a dynamic updating mechanism, reply of image-text audio-video combination is achieved, and the fact accuracy is guaranteed through node association priority matching; the system explicitly guides the model reasoning process, establishes a traceable logic chain, effectively reduces knowledge illusion while improving the answer real-time performance, and has the characteristics of multi-round dialogue continuity and humanized interaction.

The invention discloses a cluster computing power energy efficiency perception scheduling and green computing system, which relates to the technical field of computers and comprises a multi-source energy efficiency perception and data acquisition module used for acquiring power consumption, utilization rate, temperature, cooling state, PUE index and environmental data of cluster nodes. According to the invention, through the multi-modal energy efficiency fusion sensing network and the multi-scale convolution and time sequence attention fusion network, multi-source heterogeneous energy efficiency data such as current, voltage, temperature, airflow and the like of a node level can be collected and fused in real time and with high precision, noise is effectively removed, abnormity self-correction is realized, the defect of energy efficiency sensing granularity in the prior art is made up, and the energy efficiency sensing precision is improved. And reliable input is provided for subsequent scheduling decisions. A cross-scale dynamic twinborn collaborative modeling mechanism is adopted, a physical information neural network and a computational fluid mechanics model are coupled, optimization is carried out through a generative adversarial network structure, and accurate prediction of a complex energy consumption evolution curve and a cooling flow field is achieved.

The invention discloses a dynamic path optimization method based on response time domain attenuation, and relates to the technical field of artificial intelligence and intelligent path planning, and the method comprises the steps: generating a node network composed of navigation points, terrain units or interaction regions, and forming a node network topology structure; generating a dynamic state feature set used for describing game scene changes, and forming input data used for follow-up node priority dynamic adjustment; converting the dynamic state feature set into node-level standardized event data, and distributing the node-level standardized event data to a node state management module through an event bus; setting a current node priority for each node in a node state management module based on the node-level standardized event data; forming a time domain attenuation model of the node priority; the priority recovery coefficient is improved; in the path planning stage, executing a dynamic path search algorithm to generate a passing path with the minimum total cost and the optimal path smoothness; when it is detected that player operation or scene change causes node response mutation, a local re-planning mechanism is triggered, smooth path transition is achieved, and overall jumping is avoided. According to the method, the problems of path congestion, frequent switching and unsmoothness caused by lack of dynamic node state and event response calculation in the prior art are solved. Through node priority time domain attenuation and dynamic path optimization, the technical effects of smooth path, efficient passing and node load balancing are achieved.

The invention relates to the technical field of wireless communication and edge computing fusion, in particular to a wireless network resource allocation method based on edge intelligence, which comprises the following steps: S1, collecting a multi-dimensional parameter set of network edge nodes in real time; s2, converting node position coordinates into a network topological structure map, and mapping service type labels and bandwidth requirements into service requirement feature vectors; s3, generating a domain division matrix based on weight distribution of the network topological structure atlas; s4, outputting a resource block configuration tensor according to the time-frequency feature of the business demand feature vector; s5, generating a resource allocation mapping relation table; and S6, generating and executing a node-level resource scheduling instruction set. According to the invention, by constructing a resource scheduling mechanism fusing a topological structure and service features, precise resource allocation and dynamic load balancing of a cooperative domain level and a node level are realized, and the efficiency and the resource utilization rate of multi-task concurrent processing in an edge computing environment are improved.

The invention discloses an equipment control method and device, electronic equipment, a storage medium and a product, and relates to the technical field of equipment management, when a master node is offline, slave nodes can evaluate own conditions, and only participate in election when state information and node levels of the slave nodes meet election conditions. In order to reflect the adaptability of the nodes to bear the responsibilities of the main nodes in real time, the equipment priorities of the nodes are determined according to the hardware resources, the performance indexes and the historical data of the nodes and the index weights matched with the current scene. According to the method, the equipment priority, the risk level, the performance index and the online duration contained in each election message in the region are analyzed to determine the supported node, so that the selected node can stably and reliably undertake the management work of the main node. And determining an election mechanism according to the number of the selected nodes, and determining the most suitable new main node. It is ensured that the new main node can successfully take over all work of the original main node, and the reliability of equipment management is improved.

The invention discloses a space-time sequence interpolation method and device oriented to heterogeneous deletion, and belongs to the technical field of space-time data processing. Aiming at random or continuous loss of the sensor network caused by faults and communication interruption, the method comprises the following steps: setting static space experts, dynamic space experts, short-term experts and long-term period experts in parallel in the same framework, and respectively capturing fixed geographical adjacency, time-varying space correlation, local continuous trend and long-period rules; spatial features are extracted through high-order diffusion diagram convolution and bidirectional gating circulation, time features are extracted through multi-layer space-time attention, a memory attention gating network is introduced to dynamically weight and fuse output of experts according to reconstruction errors, and node-level and time-step-level self-adaptive interpolation is achieved. Experiments show that compared with the prior art, the method has the advantages that under various real data sets and heterogeneous missing scenes, the precision and robustness are remarkably improved, and the method can be widely applied to scenes needing high-integrity spatio-temporal data, such as intelligent transportation, air quality monitoring and energy internet of things.

The invention discloses a multi-edge-node collaborative industrial energy consumption data compression and storage method and system, particularly relates to the technical field of data compression and storage, and realizes hierarchical compression processing by layering industrial energy consumption data acquisition equipment into a main node, a core slave node and a terminal slave node. According to the method, various compression algorithms are used for carrying out lossy or lossless compression on different types of data, and cooperative compression and data redundancy elimination between edge nodes are realized under the coordination of a main node by introducing a distance calculation mechanism of'node hierarchy + auxiliary parameters'. The system constructs a node correlation graph through data abstracts, optimizes a data transmission path based on reinforcement learning, and adopts a cold and hot data classification strategy to improve the storage efficiency. The data compression efficiency and reliability are improved, and the method is suitable for real-time processing and long-term storage of large-scale industrial energy consumption data.

The invention relates to a side cloud collaborative adaptive workflow scheduling method and system, and belongs to the technical field of distributed computing and artificial intelligence. The method comprises the following steps of: firstly, in a macroscopic candidate screening stage, reducing problem granularity through task clustering, and obtaining balance between utilization and exploration based on a weighted distance probabilistic preferential strategy; then, in a collaborative scheduling decision-making stage, a global network state diagram is constructed through a graph neural network, deep spatial features of nodes and neighborhoods of the nodes are extracted, context-aware state representation is formed, and a reinforcement learning agent makes an optimal collaborative decision in multiple options such as local execution, edge migration or cloud unloading according to the state representation; and finally, in a local adaptive optimization stage, performing fine-grained optimization after the task is issued, dynamically adjusting a scheduling frequency and a multi-target weight through an online learning mechanism, realizing balance between a task deadline and a resource utilization rate, and ensuring efficient and robust execution of a node level.

One or more systems, devices, computer program products and / or computer-implemented methods of use provided herein relate to predicting a optimized result for a GNN. A system can comprise a memory configured to store computer executable components; and a processor configured to execute the computer executable components stored in the memory, wherein the computer executable components can comprise a convolution aggregation component that aggregates features through incoming edges, and aggregates features through outgoing edges to learn the different roles of those edges. Also a node level aggregation component that aggregates edge attributes and node features from neighborhood nodes and edges to optimize computing workload and memory, and a generating component that generates a prediction for an input graph, or for a part of a graph such as a node (vertex), an edge, or a subgraph, based on the node attributes (features) and edge attributes that are attached to the input graph.

The invention discloses an FPGA (Field Programmable Gate Array) netlist-level hardware Trojan horse detection method based on a large language model, and relates to the technical field of integrated circuit safety and hardware Trojan horse detection, and the method comprises the following steps: constructing an original FPGA netlist into a text attribute graph containing textualized attributes, generating a path text sequence through bidirectional random walk, and carrying out two-way random walk on the path text sequence; constructing a corpus to pre-train a large language model; then, delimiting a local neighborhood by taking each node as a center, generating a path text set, extracting semantic vector representation of the path text set by utilizing a pre-training model, and further constructing a node-level training sample set; on the basis, supervised fine tuning is carried out by combining a classifier and CB-Focal Loss, and a final model is obtained; in the reasoning stage, representation construction and discrimination are carried out on nodes to be detected, and node-level hardware Trojan horse detection is achieved. According to the method, circuit topology and semantic information can be reserved at the same time, node-level hardware Trojan positioning is achieved, the detection precision and generalization ability are improved, and the automation degree is improved.

The invention discloses a license authorization management method for a software platform, which comprises the following steps of: storing files or data by adopting a multi-way tree type storage structure, granting a user permission of a corresponding node level according to different nodes of the multi-way tree type storage structure, generating a unique machine code according to the user permission and server hardware information, and storing the unique machine code into a server; performing digital signature on the machine code to generate a legal authorization file; finally, when the user accesses the corresponding node, performing authorization file verification, and judging whether the authorization file verification is passed or not; and the effect of browsing, viewing or modifying hierarchical management aiming at users with different permissions is realized.

The invention provides a high-performance virtual scrolling method and system supporting a large-scale tree structure, and the method comprises the steps: constructing a dual-mapping data table structure of the tree structure, which comprises a node mapping table and a sublevel relation mapping table; when it is detected that the node height changes, the virtual rolling rendering engine generates a flattened rendering list in real time based on the unfolding state of the corresponding node in the node mapping table; the virtual rolling rendering engine calculates rendering positions of nodes in the visual area through a dynamic position prediction algorithm; in response to the rendering location and the device type, the virtual scrolling rendering engine dynamically adjusts the virtual scrolling window and the tree structure node hierarchical indentation pitch, and renders in the virtual scrolling window according to the flattened rendering list. According to the application, the data relationship and the rendering position are decoupled through the double-mapping data table structure, the smooth rolling of ten thousand-level nodes can be realized, the memory occupancy rate is reduced, the interaction response speed and the rendering performance are improved, and the method has rolling stability, cross-device display consistency and deployment flexibility.

The invention discloses a topology anti-error risk management and control method and system, and relates to the technical field of risk management and control, and the method comprises the following steps: obtaining a power grid topology structure chart, and employing a graph algorithm to code power grid nodes in the topology structure chart, and obtaining a node feature vector; constructing a truth-supplementing parameter group based on the measurement missing node identification result and the node level, and performing random disturbance on the truth-supplementing parameter group based on a preset first constraint condition to obtain a plurality of truth-supplementing schemes; repairing the node feature vectors of the measurement missing nodes by adopting a truth-supplementing scheme to obtain a node repairing feature vector set, and performing similarity analysis on the node repairing feature vector set and a preset risk scene; and calculating an optimal similarity threshold value corresponding to the authentic scheme based on a similarity analysis result, constructing an optimal pairing group, and applying the optimal pairing group to actual anti-error risk identification, so that the problem that the accuracy of power grid risk management and control is reduced due to the lack of a dynamic adaptation mechanism between the repaired node features and an actual risk scene in the face of node measurement deficiency is solved.

The invention relates to a programming review method and system based on multi-modal data. The method comprises the following steps that source code fragments are collected, character format matching is carried out, consistent items are output, and initial node mapping is generated; extracting a pointer reference and calling sequence, carrying out node association recombination, screening node levels, carrying out semantic mark matching, summarizing a connection path, checking a code security scanning index, and generating a dynamic semantic structure; carrying out stack-up matching to compare pixel distribution and label similarity, scanning edge coordinates, checking label corresponding positions and visible area distribution, judging feature co-occurrence, and generating multi-modal associated features; and performing program logic checking, analyzing a variable initialization sequence, generating a checking element set, comparing code security scanning indexes, generating security scanning information, judging a risk identification level and generating a review instruction list. Compared with the prior art, the information utilization rate and the correlation analysis depth in the code review process can be improved, and the accuracy of program defect and risk discovery is enhanced.

The invention discloses a service abnormity monitoring method and system based on dynamic graph calculation, and relates to the technical field of service monitoring data processing, and the method comprises the following steps: generating a service graph structure based on real-time service interaction data and service association attributes of an ERP report, performing node rationality dynamic evaluation based on the node basic stability parameter and the associated service parameter, and executing node dynamic adjustment based on the node update resource occupation condition and the report granularity mode according to a node rationality dynamic evaluation result; performing node weight dynamic processing based on the service interaction parameter of each associated service of the node and the node rationality dynamic evaluation result, and performing service influence dynamic evaluation based on the node weight dynamic processing result and the node level; and performing node risk dynamic assessment based on the service influence dynamic assessment result and the node dynamic stability parameter, and performing monitoring dynamic regulation and control processing based on the node risk dynamic assessment result.

The invention discloses a multi-level two-state content management method and system based on an unreal engine, and the method comprises the steps: defining a level identification rule, employing a point separator to represent the father-son relationship between levels, dynamically generating tree-shaped nodes based on the level identification, and enabling each node to have corresponding node information, the node information comprises a hierarchical identification character string or a GameTag identification state provided by an unreal engine, an activation state of a current node, a global state mark of a child node and a child node reference list; and generating or matching nodes step by step by recursively analyzing a hierarchical identification character string or a GameTag identification state provided by an unreal engine to form a complete tree hierarchical structure. According to the method and the system, efficient management and dynamic expansion of multi-level double-state contents are realized, cross-project multiplexing is supported, and the method and the system are suitable for digital twin application development in a complex service scene.

The invention discloses a time sequence semantic attention-based self-supervised spatio-temporal data completion method and system, and the method comprises the steps: obtaining spatio-temporal data, and dividing an incomplete time sequence into semantic segment units through time sequence segmentation modeling; mining an inter-segment potential time sequence dependence mode by using a cross attention mechanism to obtain a segmented attention calculation result; constructing a time sequence comparison model based on a segmented attention calculation result, and obtaining strong generalization embedding of time sequence modeling through a self-supervision method; and carrying out spatial domain modeling through the graph convolutional network according to the time sequence characteristics which are obtained by time sequence modeling and are embedded as each node, so as to obtain a spatial domain modeling result. According to the method, strong dependence of a traditional model on data integrity is broken through, nonlinear space-time correlation can be effectively captured in a data missing scene, and a high-robustness time sequence embedding basis is provided for subsequent space-time node level tasks.

The invention discloses a multi-mode local voltage control method and system based on a deep neural network, and relates to the technical field of power system operation optimization, and the method comprises the steps: constructing a multi-mode sample set comprising node voltage, photovoltaic output and load data, setting a voltage upper limit, a voltage lower limit and a hysteresis interval, and dividing operation modes; establishing a control mode selection unit by using a deep neural network, and establishing a reactive power output prediction model by using a convolutional neural network; and performing localized deployment based on the trained controller model, and performing node-level adaptive voltage control and real-time reactive power optimization according to the hysteresis interval and the residence time. According to the method, stable division and feature layering of voltage states are achieved, representativeness of model training samples and precision and stability of voltage mode recognition are improved, collaborative optimization of mode judgment and adjustment output is achieved, stable switching and dynamic optimization of the voltage states are achieved, and system immunity and stability of multi-node operation are improved.

The invention relates to a desensitization method, system, equipment and medium for industrial data, and the method comprises the steps: building and dynamically maintaining an industrial field standard data variable rule table, and defining a desensitization strategy and a sensitivity level matched with specific industrial variables such as temperature, rotating speed and the like in physical meanings, time scenes and node levels of the specific industrial variables; the method comprises the following steps: adding a multi-dimensional identifier to original industrial data, establishing a six-dimensional binding relation model of data variable type-time dimension-node hierarchy-cluster role-sensitive level-rule entry, driving the data to execute a time-coordinated stepped desensitization process in four node hierarchies of a workshop, an enterprise, a park and a cluster, according to the method, refined access control and data decryption fusing time, space and role dimensions are realized according to the model and a full-link tracing log, and the problems that rules and industrial scenes are disjointed, dynamic time adjustment is lacked, cross-node collaboration is insufficient and authority control coarse granularity is caused in the prior art are effectively solved.

The invention discloses an underwater Internet of Things destroy-resistant link optimization method combining self-guided graph representation and AMF. The method comprises the following steps: 1) generating a scale-free topology and a candidate link set according to node positions, residual energy and underwater acoustic channel quality; 2) inputting the topology into a self-guided graph representation learning model, and applying contrast constraint through unsupervised training to obtain robust node embedding representation; 3) on the basis of candidate link two-end embedding, combining node degree, betweenness centrality, bridge edge coefficient and link success rate to construct feature representation, and generating an existence probability matrix through a probability mapping function; 4) inputting the matrix into adaptive multi-layer filtering (AMF), reserving edges according to Top-k to form a connected skeleton, setting a hierarchical threshold according to node hierarchy and load, and maintaining approximate power law degree distribution; and 5) applying a global connectivity constraint, and outputting an optimized link structure. According to the method, link optimization can be realized without supervision labels, network connectivity and survivability are improved in an attack environment, and the method is suitable for ocean monitoring and security scenes.

The invention relates to the technical field of communication, in particular to a controller communication management system and method based on star flash, and the system comprises a protocol adaptation module, a node dynamic networking module, a data flow optimization module, a behavior pattern matching module and a risk early warning output module. And arranging communication frame timestamps through a time axis to extract a high-risk area, identifying a signal intensity reversal point and a networking abnormal area, analyzing a data flow disturbance inflection point sequence, matching a behavior pattern, and finally generating a communication risk signal set. According to the method, quantitative analysis can be carried out from a multi-dimensional structure, signal intensity, a dynamic path and a time sequence behavior, full-process closed-loop tracking and node hierarchical positioning are realized, the accuracy and timeliness of communication risk prediction are improved, and accurate logic support is provided for high-risk evolution signals.

The invention discloses an efficient storage method and system for distributed molecular dynamics simulation calculation, and the method comprises the steps: constructing a distributed storage model based on data access domain division, and dividing the data stored in a node into private access data and shared access data; a storage structure with particle multi-dimensional information separation is adopted, particle data is divided into static data and dynamic data, and different storage structures are adopted for different types of data. A hierarchical data mapping and addressing method is adopted, storage and access of particle data are divided into a simulation space level, a calculation node level and a space grid level, each layer is organized through a specific numbering rule, and effective distribution and efficient access of the particle data are ensured. According to the efficient storage method and system for distributed molecular dynamics simulation calculation, the problems of uneven particle data distribution and low access efficiency in the molecular dynamics simulation process are solved.

The invention discloses a time sequence root cause analysis method and system based on dynamic causal inference, and belongs to the technical field of industrial system fault diagnosis and intelligent operation and maintenance. The method comprises the following steps: collecting and preprocessing multivariable time series data in an industrial system; performing dynamic causal topology discovery by using an improved DAG-GNN algorithm, and generating a dynamic causal topology reflecting system state transition; performing spatio-temporal feature aggregation of causal guidance through GCN and LSTM, and extracting joint representation fusing node local abnormal information and a system-level global fault trend; node-level and system-level collaborative analysis is carried out through a multi-granularity root cause inference module, and a root cause candidate set with confidence ranking is generated; according to the method, the defects in the aspects of dynamic causal modeling, feature coupling and multi-granularity reasoning in the prior art are overcome, accurate tracing of a cross-level fault chain can be achieved, and reliable and explainable decision support is provided for intelligent operation and maintenance of an industrial system.

Methods, device and data stream are provided to adapt the conditions under which the visibility of an object triggers or does not trigger actions in an extended reality application. A visibility trigger is activated when the objects described in the nodes linked to the trigger are visible from the point of view of a given camera. Additional information is set at node level to adapt and specialize criteria of visibility triggers. This information relies on the nature and / or the size of the object geometry and on the nature and / or the size of the other nodes' geometry considered for the visibility estimation of the object.

The invention discloses an AGV full-life-cycle intelligent operation and maintenance method based on a twin map and reinforcement learning, and belongs to the technical field of equipment operation and maintenance management, and the method comprises the steps: employing an edge-level dynamic attenuation function and a node-level economic weight operator for aggregation, and obtaining a health vector containing a whole vehicle risk score and a part first-level risk identifier; the method comprises the following steps: training a reinforcement learning agent according to a three-section reward function comprising production line instant income, health improvement income and future risk deduction, and determining an operation and maintenance strategy weight file by adopting a profit-oriented gradient update mode optimization strategy with risk suppression factors; and carrying out equipment operation and maintenance management on the AGV in workshop operation according to the strategy weight file. According to the method, dynamic risk propagation and economic quantitative evaluation are adaptively fused under the real-time working condition of a workshop, and a comprehensive optimal decision of maintenance opportunity and production line income is realized under the condition of non-stop observation through a reinforcement learning strategy closed-loop driving AGV fleet preventive maintenance and scheduling method.

The invention provides a full life cycle carbon emission evaluation method based on a transformer substation prediction model, and the method comprises the steps: building an operation structure tree based on the distribution characteristics and function relevance of a physical entity object of a transformer substation by taking a carbon emission entity object of each full life cycle stage as a tree node; mapping physical attributes and dynamic data of each carbon emission entity object to a digital twin platform based on node levels in the operation structure tree to obtain virtual nodes; associating the dynamic data of the virtual node with the static data of each full life cycle stage based on the node identifier in the running structure tree to obtain a full life cycle data chain; based on the associated data of each tree node in the full-life-cycle data chain, calculating a carbon emission result in each full-life-cycle stage; and according to a father-child node relationship of the running structure tree, upwards aggregating carbon emission results of lower-level nodes of each tree node to obtain a total carbon emission evaluation amount. According to the invention, the accuracy of a carbon emission evaluation result is improved.

The invention discloses a node state extracting, triggering and uploading method and system for a substation monitoring system, and an equipment storage medium, and the method comprises the following steps: extracting function nodes and monitoring elements of substation monitoring system equipment, and constructing a general mechanism logic model in an XML format; based on node levels in the mechanism logic model and bound monitoring elements, state quantity, analog quantity and counting quantity data in the equipment are extracted, and an abnormal trigger signal is generated according to a preset trigger rule; the trigger signal and the self-checking result are converted into a CIM / E format log file, and the CIM / E format log file is stored and uploaded to an operation and maintenance platform based on a log rotation mechanism; according to the invention, more comprehensive and general data support is provided for scheduling control and equipment operation and maintenance.

Methods, device and data stream are provided to initiate an extended reality application that uses a physics engine. The scene description comprises behavior data based on triggers. At least one trigger is a collision trigger that is activated in case of a collision between two objects described in nodes of the scene graph of the scene description. The scene description comprises, at the node level, a description of how the object of a given node has to be considered by the physics engine in order to optimize the use of memory and processing resources. A node is of one of three categories: to be ignored by the physics engine, to be considered only for collision or to be considered for a full physics simulation. At the starting, the physics engine is initialized according to these categories and the extended reality application is started in the best conditions.

Embodiments of the present disclosure provide a data processing method and cluster, a computing device, a computer readable storage medium, and a computer program product. The data processing method is applied to an application program interface service unit of the data processing cluster. The method comprises: receiving a data processing request sent by a client for a target storage unit, wherein the data processing request carries data to be processed and a unit path of the target storage unit; analyzing the data to be processed, and when it is determined on the basis of an analysis result that the data attribute of the data to be processed is a target data attribute, encrypting the data to be processed to obtain ciphertext data and an encryption key; and sending the ciphertext data and the encryption key to the target storage unit on the basis of the unit path, thereby minimizing an attack surface of a malicious user to a node level, and reducing the risk of sensitive data leakage.

The invention provides an attack detection and source tracing method and system for a multi-mode AI system, and relates to the technical field of artificial intelligence. The method comprises the steps that input data, model internal states, system logs and behavior data are collected, and time mark high-frequency records are unified to form a multi-modal data set; encoding each time window graph snapshot of the dynamic graph by applying a neural network, and generating a graph overall embedded vector and a node embedded vector to capture spatial structure and time sequence dependence; acquiring a graph embedding representation of a current time point according to the graph overall embedding vector and the node embedding vector, executing attack detection based on the graph embedding representation, and triggering an alarm when an anomaly is detected; attack traceability is carried out in an abnormal time point diagram snapshot, a node-level abnormal score is calculated through node embedding, and key nodes, edges and sub-graphs are identified by using attention weight of a neural network and a graph interpreter so as to locate attack entry points, propagation paths and influence ranges.