292 results about "Central node" patented technology

The invention relates to the technical field of micro-grid energy optimization, in particular to a distributed federal reinforcement learning-based hydrogen-containing micro-grid energy scheduling method, which comprises the steps of constructing a multi-region hydrogen-containing micro-grid system model, designing a state space, an action space and a reward function of an intelligent agent, constructing an Actor-Critic network and an experience pool, and completing environment initialization. The intelligent agent inputs the operation state of the equipment into the Actor network, updates the state of the equipment according to the output action, verifies the constraint and outputs a reward value; tuples are extracted from the experience pool to update local network parameters, and the exploration rate is updated regularly; when a federation interaction period is reached, exchanging Critic network parameters and updating federation parameters; and when the training round arrives, outputting an equipment operation plan, deploying the model to the local hydrogen-containing micro-grid in the island mode, and outputting an equipment output value. According to the scheme, strategy sharing and learning collaboration are realized through neighborhood collaboration and local communication among the regional intelligent agents, so that dependence on a central node is avoided.

The invention discloses an agent collaborative generation method based on dynamic sharing of bureau-school interconnection teaching resources, which comprises the following steps: S1, constructing a bureau-school interconnection network system, and deploying a resource management agent which comprises a data acquisition module, a semantic analysis module, a decision module and a communication module; s2, collecting school teaching data, and sending the school teaching data to a central node of an education management department; s3, performing semantic annotation and classification on the received data, and constructing a dynamic teaching resource semantic database; s4, constructing an intelligent agent collaborative decision-making model, and dynamically adjusting a sharing strategy of teaching resources; s5, generating an optimal resource sharing scheme through the dynamic teaching resource semantic database and the intelligent agent collaborative decision-making model, and sending the optimal resource sharing scheme to the school child node; and S6, after sharing is completed, a resource sharing scheme is fed back to the center node, and the agent collaborative decision model is optimized. According to the invention, intellectualization, dynamics and precision of teaching resource sharing are realized, and a brand new technology is provided for balanced configuration of education resources.

The invention is suitable for the technical field of unmanned aerial vehicle cluster cooperative control, and provides an edge calculation distribution system and method for unmanned aerial vehicle cluster cooperative tasks, and the system comprises a multi-heterogeneous neural network module, a local fusion module, a global fusion module, a conflict resolution module, a task distribution module, and a self-adaptive optimization module. According to the invention, by constructing a multi-level semantic feature extraction and fusion architecture and combining an intelligent conflict resolution mechanism and a dynamic task allocation algorithm, efficient intelligent collaborative operation of an unmanned aerial vehicle cluster in a complex and changeable environment is realized, all core algorithms can be executed in a distributed manner in the local of the unmanned aerial vehicle through a distributed edge computing architecture, and the distributed edge computing architecture can be applied to the unmanned aerial vehicle cluster. Dependence on a central node is remarkably reduced, communication overhead and delay are greatly reduced, cooperative work can still be maintained even if part of nodes fail, and perfect balance between robustness and performance is achieved.

The invention belongs to the technical field of data processing, and provides a data management optimization method and system. The method comprises the following steps: based on an acquisition strategy, acquiring log and business data of each child node, and calculating a data acquisition success rate and a data acquisition delay evaluation value to measure an acquisition condition; transmitting the collected logs and service data of each child node to a central node, carrying out multi-class data processing, and calculating and determining a quantitative index of a data processing delay condition so as to optimize a data processing process; monitoring the data after processing the multiple types of data in real time, identifying a risk event by using a self-constructed risk event identification model, and calculating a risk score of the risk event, including analyzing and calculating a risk monitoring result so as to update a monitoring rule and a feature library of the risk event identification model; and when the calculated risk score exceeds a preset threshold value, an alarm mechanism is triggered, and alarm information is pushed to safety management personnel of the related child nodes. According to the invention, the multi-node data acquisition and data processing process is effectively optimized.

The invention discloses a power distribution network short-circuit parameter real-time monitoring method, system and equipment based on edge calculation and a medium, and belongs to the technical field of power system monitoring and protection, and the method comprises the steps: collecting data information in real time at each monitoring node of a power distribution network, generating a state vector reflecting a node operation state, and inputting the state vector to an edge calculation unit, processing through an internal first calculation model to obtain a power grid fault evaluation result of a corresponding node, and compressing and uploading the power grid fault evaluation result to a cloud; based on the uploaded data and the system operation state, a system performance evaluation result is generated in real time, and an adaptive optimization algorithm is used for dynamically adjusting internal model parameters of the edge calculation unit. According to the method, the composite algorithm model fusing feature extraction and parameter calculation is deployed at the edge node, the problems that a traditional centralized processing mode is large in data transmission delay and central node calculation pressure is concentrated are solved, and the technical effect of reducing bandwidth occupation of a backbone communication network is achieved.

The invention discloses a hybrid federal learning modeling method and device for a neurocognitive impairment evolution trajectory, and the method comprises the steps that a child node receives an expert global model distributed by a central node, and the expert global model is determined by the central node from an expert global model group according to metadata pre-registered by the child node; taking the expert global model as an initial model, and performing localization training by using the local multi-modal non-invasive physiological signal data to obtain a trained local model; extracting global sharing parameters in the trained local model, and uploading the global sharing parameters to a central node; and for the corresponding expert global model, the central node aggregates global shared parameters uploaded by all the child nodes training the expert global model to obtain an updated corresponding expert global model, issues the updated expert global model to the child nodes, and circulates until a termination condition is met. According to the method, the heterogeneous data of the child nodes are adapted, privacy is protected, and the adaptability and accuracy of trajectory prediction are improved.

The invention provides an intelligent question answering method and system based on a traditional Chinese medicine classical meta-term engine. The method comprises the following steps: identifying and extracting core traditional Chinese medicine terms and a user query intention in a user question; the core traditional Chinese medicine terms are input into a pre-constructed traditional Chinese medicine classical meta-term engine to be processed, normalized mapping and semantic association extension of the terms are completed, and the engine is constructed based on a classification-term-word list-semantic CTVS association framework. Comprising a traditional Chinese medicine classical book term ancient and modern double-classification system C layer, a meta-term extraction and standardization T layer, a dynamic word list V layer and a semantic association network S layer. Carrying out graph pattern matching and reasoning in the S layer, traversing and integrating a plurality of deep association paths by taking classical element terms and corresponding modern clinical terms as double center nodes, and retrieving answer information associated with the query intention; and carrying out integration sorting and structured assembly on the answer information to generate a structured answer. Intelligent questions and answers can be performed on traditional Chinese medicine related questions of the user.

The invention discloses a hybrid cloud-oriented multi-cluster scheduling method and device and a medium, and relates to the technical field of distributed system resource management. The method comprises the following steps: continuously collecting multi-dimensional state data of a cluster to be scheduled, and encrypting and transmitting the multi-dimensional state data to a central node to obtain a global state view; if the multi-dimensional state data meets a preset triggering condition, analyzing a service strategy associated with the global state view to obtain a structured constraint condition and weight configuration; removing clusters contained in the global state view based on a structured constraint condition to obtain a candidate cluster list; calculating a cost factor, a performance factor, a utilization rate factor and a stability factor of each candidate cluster in the candidate cluster list, and performing weighted calculation with the weight configuration to obtain a comprehensive score of each candidate cluster; and extracting the cluster with the highest comprehensive score as a target cluster, and calling the integrated scheduling executor to perform flow scheduling and resource scheduling on the target cluster.

The invention provides an equipment networking method and system based on a distributed soft bus, and the method comprises the steps: starting a distributed soft bus module and carrying out the discovery of adjacent equipment, and obtaining a trusted equipment set after identity verification; a point-to-point communication link between the devices is established according to the trusted device set, and a distributed mesh interconnection topology is formed; performing routing calculation and data forwarding on data transmission in the mesh topology to realize end-to-end communication; meanwhile, dynamic changes of equipment in the network are monitored, topology updating is carried out, and the network stability is maintained. The scheme does not depend on the dependence of the traditional star network on the central node, realizes direct interconnection between devices through the distributed soft bus technology, improves the fault-tolerant capability and transmission efficiency of the network, and solves the problems of poor network reliability and limited expansibility in the prior art.

The invention discloses an interpretable clinical decision support system based on label generation and a knowledge graph. The method comprises the following steps: based on a breast cancer domain knowledge enhanced version Qwen-BrCaAdapt of a general large language model Qwen, analyzing an unstructured medical record text of a patient, and generating a structured result containing tags, values, evidences and explanations; calculating a reasoning label through a path matching engine by utilizing an editable structured path rule table, and matching a candidate treatment scheme according to the reasoning label; and taking the matched treatment scheme as a central node, calling a medical knowledge graph to bind entity information including clinical evidence, recommendation levels, medical insurance information, medication risks, usage and dosage, and generating a traceable JSON structure and a visual report. The method has the beneficial effects that the accuracy and efficiency of tag generation in the breast cancer field are improved, the rule maintenance cost is reduced, the interpretability and traceability of a clinical decision scheme are enhanced, and the acceptability of a doctor to a recommendation result is improved.

The invention discloses a global unique identifier generation method and system under edge computing, belongs to the technical field of edge computing, and aims to solve the technical problem of how to generate a unique identifier under edge computing according to the environmental characteristics of edge computing, and the method has high efficiency, uniqueness, incremental performance and safety at the same time. According to the technical scheme, the method comprises the following steps: applying for a unique identifier: an edge node sends a node ID of the edge node to a center node, and further applies for a global unique identifier from the center node; calculating a timestamp: after the center node receives the application of the edge node, acquiring the current accurate time, and calculating the timestamp with the length of 40 bits; determining a service number: directly using the ID submitted by the edge node as a 12-bit service number by the center node; acquiring and progressively increasing a serial number: acquiring the current serial number from the database by the central node, and progressively increasing the current serial number by 1 to obtain a 12-bit serial number; encrypting the confused ID; and returning the unique identifier.

The invention relates to a task scheduling method and system, electronic equipment and a storage medium. The method comprises the following steps: respectively acquiring task attribute information corresponding to a plurality of to-be-processed tasks, and respectively acquiring resource attribute information corresponding to a plurality of edge nodes; based on the task attribute information and the resource attribute information, distributing the plurality of to-be-processed tasks to corresponding edge nodes for processing; in the task processing process, if it is monitored that the resource utilization rate of any edge node is overloaded, other edge nodes or center nodes are called for cooperative processing. According to the scheme, the task execution efficiency, the resource utilization rate and the task scheduling efficiency can be improved.

The invention discloses a high-dimensional data retrieval method based on distributed computing and mixed indexing. The method comprises the following steps: step 1, data preprocessing; step 2, cluster generation; step 3, index construction; step 4, consistent Hash: distributing each cluster to a virtual node of a Hash ring, dynamically adjusting data distribution on the virtual node according to a load strategy, and mapping the virtual node to a physical server; and step 5, hierarchical aggregation: mapping a query request to a node and adjacent nodes on a hash ring through a hash function, performing candidate set calculation in parallel, merging candidate sets sent by all the nodes by a Spark center node, and finally optimizing a result. According to the method, high-recall-rate and low-delay distributed retrieval is achieved through mixed index construction, dynamic consistency hash allocation and hierarchical aggregation strategies, meanwhile, system load balancing and index updating efficiency are guaranteed, and the method is suitable for large-scale high-dimensional data scenes.

The invention is suitable for the field of power systems, and provides a power grid attack defense method, terminal equipment and a storage medium. The power grid attack defense method comprises the following steps: an edge node determines local abnormal information and a first response strategy according to local communication data, wherein the local abnormal information represents power grid attack information suffered by the edge node; determining local association information according to the local exception information and shared exception information of the adjacent nodes, wherein the local association information represents an association relationship between the edge node and the plurality of adjacent nodes; sending the local association information to the central node; obtaining traceability information obtained by the central node according to the local association information, wherein the traceability information represents attack source information and attack path information of the power grid attack; and determining a second response strategy according to the traceability information and the local abnormal information. Through the synergistic effect of the edge nodes and the center node, the full-chain identification and multi-point protection capabilities of the power system in a complex attack environment are improved.

The invention discloses a collaborative awareness and recognition method and system based on an unmanned aerial vehicle cluster, and relates to the field of unmanned aerial vehicle cluster control and computer vision. The method comprises the steps of starting ground control software to load a route path file, constructing a sequence image optical scene and evaluating the effect, dynamically fusing and evaluating the effect of a wide-area optical scene, intelligently and collaboratively recognizing an optical target, confirming and judging collaborative recognition data, and counting and evaluating recognition accuracy. The system comprises a central node unmanned aerial vehicle subsystem, an edge node unmanned aerial vehicle subsystem and ground control software. According to the method, the central node and the edge nodes cooperatively work, the global scheduling decision-making capability of the central node and the real-time computing capability of the edge nodes are exerted, efficient coverage of a wide-area scene, high-resolution image splicing and cooperative recognition and tracking of a dynamic target are achieved, the problems of unbalanced computing load, information redundancy and the like of a traditional architecture are solved, and the method is suitable for large-scale popularization and application. The image quality and the target recognition accuracy are improved, the reliability is high, and the expandability is high.

The invention discloses a multi-cluster hot debugging method based on Service Mesh, and belongs to the technical field of cloud native software development. Through a debugging instruction, a multi-cluster intelligent routing gateway establishes a unified and logically isolated debugging session and a virtual network environment for a developer terminal and a plurality of target Kubernetes clusters; the gateway dynamically allocates a local port for a debugging task from a global resource pool, and instructs a target cluster to create a Pod; the gateway serves as a center node, performs cross-cluster routing conflict detection, generates a globally consistent dynamic routing rule and issues the dynamic routing rule to a target cluster; and finally, forwarding the online traffic matched with the rule to an appointed port of a developer local machine through a proxy Pod and a gateway. Through a centralized intelligent coordination architecture, various conflicts in a multi-cluster environment are fundamentally avoided, and the debugging efficiency, stability and automation level are remarkably improved.

The invention relates to the technical field of data processing, and particularly discloses a discrete data collection method and system based on hierarchical aggregation, by constructing a hierarchical aggregation architecture of computing nodes, intermediate aggregation nodes and a central node, the computing nodes cache discrete data to a local pre-write log queue, and the central node stores the discrete data in the local pre-write log queue; when a time window or a data volume threshold value is met, the data are subjected to batch processing compression and then sent to the corresponding intermediate aggregation nodes, the intermediate aggregation nodes decompress the data, aggregate and deduplicate the data and send the aggregated data to the center node through a pre-write log and a check point persistence state, and the center node can achieve non-repeated write-in of the aggregated data; and a back pressure signal can be sent to adjust the transmission rhythm of the intermediate aggregation node, and data collection is ensured to meet accurate primary semantics through cooperation of a transaction ID, a check point and an idempotent interface. The method can eliminate the I / O bottleneck of the central node, improves the collection efficiency and data reliability, enhances the fault-tolerant capability, and is suitable for discrete data collection of a large-scale distributed system.

The invention discloses a federal learning acceleration method, system and device based on an auto-encoder and residual quantization. A central node starts a federal learning task and distributes an initial global model, each participant node performs initial training by using local data, and an encoder and a quantizer are not used in the stage. Meanwhile, the center node trains an auto-encoder and a residual quantizer in parallel, the trained auto-encoder and residual quantizer models are distributed to each participant node, model parameters are compressed by using the auto-encoder, the compression precision is improved through a residual quantization technology, the center node collects compressed and quantized parameters, and the compression precision is improved. And reconstructing and aggregating by using a decoder, distributing the updated global model to each node for the next round of training, and repeating the process until the preset number of times of training rounds is reached or the performance of the model meets the requirement. Through the auto-encoder and the residual quantization technology, the data transmission quantity is remarkably reduced, model training is accelerated, and federal learning efficiency and practicability are improved.

The invention relates to the technical field of knowledge graph vectorization, and discloses a medical field-oriented knowledge graph vectorization representation method, which comprises the following steps of: firstly, constructing a multiple heterogeneous topological view based on a heterogeneous knowledge graph, and generating a detection parameter set covering a global fusion space so as to construct joint observation data corresponding to different distribution modes; then establishing a joint reconstruction task containing a public potential semantic base, and promoting a public base matrix to aggregate global structure information of all observation data by using an iterative algorithm; and finally, analyzing the converged model to extract a center node vector and generate an index.

The invention relates to a federated learning-based privacy protection personalized recommendation system, which comprises a user side, an edge side and a cloud side, and is characterized in that the user side is a basic unit for data generation and local training, each user maintains a personalized recommendation model on own local equipment, the edge side is used as an intermediate aggregation layer, and the cloud side is used as a cloud side; the cloud end is used for preliminarily aggregating local models of users in a certain geographic area or logic area, the edge end is generally deployed in an edge server or an area data center, and the cloud end is a central node for updating and distributing a global model and is used for secondarily aggregating area aggregation models uploaded by the edge ends to form a global recommendation model; according to the scheme, user privacy is strictly protected, and collaborative recommendation is realized under the condition of not transmitting original user data through a differential privacy and security aggregation dual protection mechanism.

The invention provides a data synchronization method, computing equipment, a storage medium and a program product, and the data synchronization method comprises the steps: detecting a connection state with a central node, and starting a local version number generator under the condition that the connection state is an off-network state; processing the target business data according to the operation instruction for the target business data to obtain an operation result of the target business data; according to the operation result and a local version number generator, operation log information generated for the operation instruction is generated and stored, the operation log information comprises an information local version number, and the information local version number comprises operation time information, an operation type and a work node identifier; and under the condition that the connection state is a recovery state, sending the operation log information of the target service data to the central node, so that the central node performs data synchronization according to the information local version number of the target service data. According to the method, an information local version number is designed, and it is ensured that data operation is traceable during network disconnection.

The invention discloses an unmanned ship cluster distributed cooperative path planning method and system, and the core of the method is that each unmanned ship intelligent body constructs a future space-time risk map through probabilistic prediction; when potential collision conflicts are detected, related unmanned ships do not need to be intervened by a central node, and finally, the party with the lower cost actively undertakes the avoidance responsibility through a distributed negotiation mechanism by bidding the multi-target comprehensive cost paid for avoiding the conflicts, so that an optimal collaborative path for the whole cluster is achieved. According to the method, through probabilistic prediction, distributed negotiation and dynamic multi-objective optimization, the safety, the adaptability and the expandability of unmanned ship cluster path planning are remarkably improved, meanwhile, the communication overhead is greatly reduced, and efficient, safe and intelligent autonomous collaboration in a complex environment is achieved.

The invention discloses a supermarket cash register data real-time processing and privacy protection method based on edge computing. According to the invention, through edge node localization training and a gradient parameter sharing mechanism, the data transmission pressure of a central node is obviously reduced, the convergence speed of a global model is effectively improved through a dynamic weighted aggregation algorithm, the iteration period of a transaction risk control model is greatly shortened, and the response delay of a system to abnormal transactions is controlled at an extremely low level; a high-concurrency transaction scene can be efficiently processed, the real-time detection capability and the system stability are enhanced, a differential encryption strategy is implemented for data with different sensitivities by a hierarchical privacy protection architecture, the unpredictability of a core data key is ensured by quantum random number encryption, a dynamic differential privacy algorithm automatically adapts to data distribution characteristics when noise is added, and the real-time detection capability and the system stability are improved. A hardware security module is combined with a threshold secret sharing mechanism, and a multi-layer protection system is constructed from a physical layer to a protocol layer, so that the risk of data leakage is effectively reduced, and meanwhile, the compliance requirements of related laws and regulations are met.

The embodiment of the invention provides a community discovery method and device for banking business, equipment, a medium and a program product, and relates to the field of big data. The method comprises the following steps: acquiring user and banking service opening information, and constructing a network model; according to the network model, calculating the local density and the segmentation distance of each node; determining a clustering center node according to the local density and the segmentation distance of each node; distributing each non-clustering center node to a community where a node which is closest to the non-clustering center node and of which the local density is higher than that of the non-clustering center node is located for the first time; for each non-enveloping node, calculating the membership degree of the non-enveloping node to a community to which the adjacent node belongs; and if the degree of membership of the non-enveloping node to the community to which any adjacent node belongs is greater than the degree of membership of the non-enveloping node in the initially allocated community, adding the node to the community to which the adjacent node belongs. According to the method, the accuracy of bank user community division is improved.

The invention discloses a distributed training method for a new energy cluster power prediction model. The method comprises the following steps: collecting time sequence data of each power station in a target range, and carrying out abnormal data detection, deletion and missing data filling operation; constructing a basic structure of a power prediction model corresponding to the edge nodes, the regional coordination nodes and the global center node in the target range, and initializing model parameters; performing regional adaptation fine tuning on training data of the edge nodes, performing iterative training on a power prediction model of the edge nodes to obtain local model parameters of each edge node, performing regional aggregation and global aggregation on the local model parameters of each edge node to obtain a power prediction model of each edge node; obtaining model parameters of power prediction models corresponding to the regional coordination nodes and the global center node; and carrying out convergence judgment on the model parameters corresponding to the global center node, if the model parameters converge, stopping iterative training, and if the model parameters do not converge, executing next iterative training, so that engineering practicability and expandability are considered.

The invention discloses an intelligent factory data monitoring method and system based on distributed collaborative sampling, belongs to the field of intelligent factory data monitoring, and aims to solve the problems of real-time monitoring, quality control, equipment maintenance and production optimization in the production process. The system is composed of a data acquisition module, a production control module, a local processing module, a data center module and a user interaction module. And efficient data processing and storage are realized through the distributed collaborative sampling network, and the expandability and flexibility of the system are optimized. The production management and control module is responsible for distribution and sampling detection of production tasks, ensures product quality and realizes production continuity when equipment is abnormal. The local processing module analyzes equipment data, establishes a prediction model, and timely finds and early warns deviation in equipment and production. The data center module analyzes early warning production data in real time, transmits an early warning signal through a central node, and drives a production line to operate. And the user interaction module provides dynamic information updating, so that a user can obtain and feed back data in real time.

The application discloses a sample data labeling system, a labeling method applied to the sample data labeling system and related devices. The sample data labeling system comprises an edge node and a center node. The edge node acquires key features of sample data, judges whether the sample data is unknown sample data according to the key features, performs labeling processing on the sample data when the sample data is unknown sample data, obtains a first labeling result, and sends the first labeling result to the center node. The center node receives the first labeling result, performs consistency processing on the first labeling result when the first labeling result indicates that the unknown sample data is successfully labeled, obtains a second labeling result, and performs labeling processing on the unknown sample data when the first labeling result indicates that the unknown sample data fails to be labeled, and obtains a third labeling result.

The invention provides a federal forgetting method and system for dynamic heterogeneous perception, and the method comprises a first training stage and a second training stage, and the data of an edge node of a target is forgotten in the second training stage; in the first training stage, federal learning is adopted to train a center node and edge nodes in the distributed network; the step of the second training stage comprises a plurality of training rounds, and each training round comprises the following steps: completing local training of the edge node by adopting training data, determining a first parameter difference based on model parameters of a local model before and after the local training, determining a contrast round of the training round in the first training stage based on a calibration interval, determining a second parameter difference based on model parameters of the local model before and after local training of the contrast round; and determining correction differences based on the first parameter differences and the second training differences, aggregating the correction differences of the edge nodes, and correcting model parameters aggregated in the training round.

According to the citation network node intra-class mixing classification method based on the node class centrality, class center sampling, intra-class mixing, neighbor selection, edge screening and adaptive loss are not isolated, an efficient cooperative enhancement chain is formed, class center nodes are screened out through construction of a multi-stage class center system, the nodes of the class centers are fused, and therefore the classification efficiency of the citation network nodes is improved. After fusion, connection is not performed on source nodes (namely parent nodes) but high-quality nodes screened out through integrated prediction consistency, then dynamic edge screening (node degrees and semantic similarity) is performed on the high-quality nodes, and through organic combination of noise suppression and structure optimization, the accuracy and generalization of citation network node classification are effectively improved; the method has wide use value and application prospect in the field of image processing.

The invention belongs to the technical field of ocean monitoring, and particularly relates to a wide-area target monitoring intelligent buoy system which comprises a water surface part, a floating body part and an underwater part. The water surface part is integrated with a visible light vision acquisition module with a rotating mechanism, a sea surface environment monitoring module, an antenna module, a positioning module, a remote sensing receiving module, a communication module and the like; the floating body part comprises a floating body block, and an embedded cluster intelligent processing module, a data storage module and a power supply module which are arranged in the floating body block and communicate with one another; the underwater part comprises an underwater acoustic sensing module and an underwater magnetic sensing module. According to the invention, through integration of a plurality of sensors, water surface and underwater integrated cooperative detection is realized; the embedded cluster intelligent processing module adopts a cluster architecture without a central node, has the capabilities of task dynamic allocation and fault seamless switching, and greatly improves the comprehensive detection performance, the intelligent decision-making level and the long-term stability in a marine environment of the system.