13 results about "Dependency network" patented technology

This invention discloses a method, apparatus, electronic device, and storage medium for cross-device NPU invocation. The method includes: converting the NPU of a terminal device within a network into an NPU service object through an NPU virtual service, wherein the number of terminal devices within the network is at least two, and each terminal device within the network possesses an NPU; registering the NPU service objects of each terminal device to the distributed hardware management service by calling the registration interface of the distributed hardware management service through the NPU virtual service; when a target terminal within the network receives a task to be processed, querying the target information of each NPU service object within the network through the distributed hardware management service, and determining the target NPU service object for processing the task based on the target information. The technical solution of this invention solves the problems of current edge-cloud combined methods for processing neural network tasks, which rely on network latency and cannot meet all operating conditions, achieving efficient utilization of the NPUs of devices within the network and improving task processing efficiency.

This invention discloses a method for mitigating social propagation risk and selecting robust groups based on extreme value dependency networks and maximally independent sets. The method includes: acquiring user interaction data from a social network and constructing a graph structure based on the user interaction data; obtaining target node pairs from the graph structure for a preset propagation activity period and calculating the extreme value dependency of the target node pairs; constructing propagation networks with target nodes as vertices and extreme value dependency as the edge construction criterion under different dependency thresholds, evaluating the propagation networks, and determining the extreme value dependency propagation network based on the evaluation results; dividing the extreme value dependency propagation network into communities and determining central nodes from the divided communities; constructing a minimum extreme value dependency group based on the extreme value dependency propagation network; and identifying high-risk node groups in the minimum extreme value dependency group using a propagation loss index to obtain the identification result. This invention can achieve the identification, isolation, and control of tail propagation linkage relationships.

This invention discloses an AVB traffic delay analysis method for TSN networks in incremental scenarios. Based on network calculus, this method overcomes the shortcomings of traditional analysis methods that rely on the global GCL scheduling table. First, the method characterizes the inhibitory effect of TT traffic on AVB flow transmission. Without parsing the full scheduling table, it constructs a fast TT arrival curve using only the maximum number and length of TT traffic slots. Second, it combines the hop-by-hop reshaping characteristics of the ATS shaper and the credit constraint mechanism of the CBS shaper to construct the arrival and service curve models of the AVB flow at the output port. Finally, it calculates the worst-case delay time at the port using the maximum horizontal distance between the arrival and service curves of the AVB flow, and accumulates the values ​​hop-by-hop along the AVB flow transmission path to obtain the end-to-end worst-case delay time. Compared with traditional network calculus methods, the incremental delay analysis method proposed in this invention does not rely on the global traffic information of the TSN network and can reduce computational complexity.

The application discloses a data governance effect evaluation method, and relates to the technical field of data governance. Firstly, the application collects evaluation index time series of a complete business cycle before and after the implementation of a governance action, and then, after discretization processing, the application screens the correlation edges between indexes based on mutual information and marks the correlation polarity, so as to construct an undirected dependence network before and after the governance. Then, the application uses conditional mutual information to complete the discrimination of the causal direction, and generates a corresponding causal graph. Finally, the application compares the edge changes of the causal graph, identifies the effect offset phenomenon between indexes, calculates the net effect value, and realizes threshold early warning. The application can accurately capture the index causal dependence structure changes caused by the governance, and improves the accuracy and reliability of the data governance effect evaluation.

This invention discloses a method and system for offline speech recognition and semantic retrieval for wind turbines using hot word caching. The method establishes a dynamic hot word cache table on the local wind turbine operation and maintenance terminal, and updates it periodically with high-frequency terms from the local knowledge base. Hot words are dynamically injected into the local language model, and their weights are adjusted according to word frequency. User voice commands are collected, and the recognized text is generated using the local acoustic model and the language model after hot word injection. The text is vectorized, and the semantic similarity with feature vectors in the pre-stored vector index of the local knowledge base is calculated. Wind turbine fault handling solutions are returned based on the similarity ranking. This invention solves the problems of wind turbine operation and maintenance relying on the network and the rigidity of offline recognition, achieving high-accuracy professional terminology recognition and intelligent semantic retrieval in a network-free environment.

The application discloses a Chang'an chain network automatic deployment method based on a Kubernetes Operator, and a core scheme of the method is to define double-layer custom resources (CRD): ChainMakerNetwork is used for declaring chain-level global configuration, and ChainMakerNode is used for declaring node-level configuration; through collaborative work of a node controller and a network controller, the node controller first generates an identity for each node and reports the identity to a resource state field; after the network controller monitors that all dependent items are ready, the network controller aggregates information to generate a global chain configuration file; the scheme mechanism breaks a circular dependency of 'node startup dependent network configuration, and network configuration generation dependent on node identity', realizes one-key and declarative deployment and full life cycle management of the Chang'an chain network in a Kubernetes environment, and greatly improves deployment efficiency, reliability and operation and maintenance automation level.

This invention relates to the field of industrial internet data intelligent processing technology, specifically an artificial intelligence-based industrial internet intelligent manufacturing data processing method. The method includes: receiving real-time sensing signal packets encapsulated with spatiotemporal markers from a manufacturing unit; parsing and separating the spatiotemporal markers from the main process state signal; retrieving and comparing historical signals with temporally adjacent signals based on the spatiotemporal markers; combining historical and current signals that meet correlation conditions into signal events and generating identifiers; extracting process state sequence fragments and aggregating them into process behavior units; calculating the state transition probability of signal events and constructing a temporal dependency network; and triggering targeted backtracking analysis of the corresponding signal event when an abnormal probability transition path is detected, generating and issuing intervention instructions for the manufacturing unit's process behavior. This method can achieve spatiotemporal correlation processing of signals and process timing modeling, accurately locate anomalies and provide targeted intervention, improving the accuracy of data processing and process control.

This application relates to the field of computer communication and security control technology, and provides a method and system for access control management of archive storage rooms based on TCP / IP communication, solving the problems of heavy network dependence and inflexible access control. The method includes: receiving an archive borrowing request from an archive management terminal; verifying whether the corresponding operation permission is within a pre-set valid time period based on the borrowing request, and generating an access validity certificate if the verification is successful; transmitting the access validity certificate to the access control controller via a TCP / IP communication network; receiving and parsing the access validity certificate, and after successful identity verification, determining whether the current time is within the time window allowed by the access validity certificate; and, based on the determination result, performing the opening or closing operation of the archive storage room door lock. The technical solution provided by this application not only achieves accurate, secure, and network-independent control of access permissions to archive storage rooms through offline verification of digital certificates.

Systems, apparatuses, methods, and computer program products are disclosed for estimating a crypto-agility score of a network of computing assets. An example method includes receiving a network graph and a set of expert-determined importance weights and selecting a first network node from a set of network nodes. The example method further includes computing an upgrade cost for the first network node based on a set of dependency network nodes and applying an expert-determined importance weight to the upgrade cost for the first network node to determine a weighted upgrade cost for the first network node. The example method further includes adding the weighted upgrade cost for the first network node to a set of weighted upgrade costs and summing, by the graph circuitry, each weighted upgrade cost from the set of weighted full upgrade costs to determine the crypto-agility score for the network of computer assets.

The application provides a software supply chain dependency network construction method for complex industrial software, and belongs to the technical field of software security.S1. main dependency analysis stage: through package manager multi-language configuration analysis, source code static analysis and construction log analysis, a complete direct dependency list is compiled;S2. metadata analysis and transmission extension stage: metadata is obtained by querying a software registry and a transmission closure is recursively constructed;S3. dependency graph model modeling stage: a directed graph model is constructed and stored in a Neo4j graph database.The application systematically solves the key bottleneck in the current technical solution, and further provides support for the analysis of the overall reliability, safety and other problems of industrial software.

The application provides a standard system model construction system and method, and relates to the technical field of data processing, and comprises the following steps: a parsing module is used for content parsing of standard files in a standard file set, and a dependency network between standards is generated; a positioning module is used for determining that two system observation points are located at an intersection core area of a business stage logical reference axis and a technical level logical reference axis in a two-dimensional logical space constituted by a logical topology skeleton, and determining that three system observation points are respectively located at left upper, right upper and lower edge logical areas of the logical space away from the intersection core area. Through the cooperation of multiple modules, the application realizes the following steps: the target industry full life cycle work range is combed, the standard dependency relationship is parsed, the observation point is positioned, the evaluation reference plane is constructed, the standard distribution is compared and analyzed, the network is corrected and integrated, and finally the standard system model covering the full life cycle of the industry and fitting the actual technical logic is generated, thereby providing comprehensive and objective support for the industrial standardization management.

The application discloses an offline automatic decryption method and device for BitLocker encrypted partition. The method extracts data required for offline DPAPI decryption and key cache data stored in a registry by loading a system disk or an image; the cache data is decrypted by using a DPAPI technology to obtain a globally unique identifier and a recovery key; the identifier is used to identify a target encrypted partition based on a preset encryption signature feature, and the recovery key is used to complete decryption and file system analysis. The application does not require a user to provide a password or rely on a network, realizes full-automatic decryption by analyzing a left cache of an automatic unlocking function, and significantly improves the efficiency of data evidence collection and recovery in an offline scene.