131 results about "Local domain" patented technology

The invention provides an adaptive path selection system based on GNN and multi-agent DRL, which is based on a layered multi-domain SDN architecture and comprises a root server layer, a domain controller layer and a data layer. The root server layer comprises a root server and a root agent, and the root agent is used for calculating a global routing path strategy and optimizing and adjusting a routing path according to feedback of a local network domain. And the root server distributes the routing path to a domain controller layer according to a routing path strategy to carry out routing path forwarding and adjustment of a local network domain. The domain controller layer comprises a plurality of domain controllers and domain intelligent agents, the domain controllers construct a topological structure composed of boundary switches, and a routing decision is made in a local domain according to an optimal global routing path strategy. And the domain intelligent agent performs modeling analysis on the topological structure by using a deep reinforcement learning algorithm in combination with a graph neural network to realize intra-domain path selection. And the data plane layer feeds back the network topology information to the domain controller, and receives and executes a routing decision sent by the domain controller at the same time.

The invention discloses a Linux container-based Android and open source gap dual-system collaborative operation method, relates to the technical field of operation system collaboration, and aims to reduce resource occupation and improve collaboration and switching coherence, and the method comprises the following steps: step 1, by taking an Android as a host, creating a container isolation running domain, starting a namespace and managing resources by a control group; the method comprises the following steps: presetting a binder virtualization and three-state template; 2, deploying an open source gap in a container, mounting a shared directory in a directory binding manner, configuring bridging or address translation, and establishing copy-on-write and snapshot identifiers; and step 3, establishing a cross-system binder channel, and setting a service registration agent in a service manager and completing mapping of a handle and an authority domain by combining a shared memory and a local socket. And step 4, executing system switching without restarting, enabling one foreground and the other background through a foreground interface and continuous quota assignment, freezing a background session, limiting the speed of background heartbeat, and returning according to a snapshot when an exception occurs.

The invention discloses a global resource-oriented cloud deck preemption control enhancement method, which relates to the technical field of resource-enhanced cloud deck preemption control, and comprises the following steps: S1, a local domain client issues a cloud deck control operation; if the cloud deck control operation is issued to the external domain client, a permission relation mapping table needs to be established; and S2, after the signaling service receives the cloud deck control operation signaling information, analyzing and obtaining information such as the equipment ID, the user identifier, the authority identifier and the domain identifier, and realizing service initialization. According to the method for enhancing holder preemption control for the global resources, the technical scheme for enhancing holder preemption control for the global resources is provided. Under important application scenes such as rapid positioning and tracking and important security, the method for efficiently enhancing holder preemption of global equipment resources is realized, normal execution of important services is ensured, and responsiveness and execution efficiency of the important services are improved.

The invention relates to the technical field of artificial intelligence, in particular to a field adaptive large model fine tuning method and system based on enterprise private data, and the method comprises the steps: extracting the feature representation of local data by each participating client through a unified pre-training model, generating a prototype set representing local region category features by adopting a mean value, clustering or random sampling method; gaussian random noise is added to the prototype set, differential privacy protection is achieved, and then the prototype set is uploaded to a server; the server summarizes prototypes of all the clients, constructs a global prototype training set, performs global training of a single communication round on the adapter module by using supervised learning classification loss under the condition that similar prototypes of different clients are not averaged, obtains a global adapter model and issues the global adapter model to the clients; and the client initializes a local model based on the global adapter model, performs fine tuning by using a small amount of data with labels, and meanwhile, prevents catastrophic forgetting through knowledge distillation constraint by taking the global model as a teacher, so that balance between global knowledge retention and local performance improvement is realized. According to the method, the problem that the expression of the model in a cross-domain reasoning task is remarkably reduced due to the fact that domain characteristics of different enterprises cannot be effectively aligned in the prior art can be solved.

The invention discloses a biomedical multi-mode signal anomaly detection and prediction method and system, and the method comprises the steps: modeling a biomedical multi-mode signal into a time-varying non-local dynamic system, capturing the long-time-history dependence characteristic of the signal through a memory mechanism of a Caputo fractional derivative, and introducing a time-varying input item to process interference; the method comprises the following steps of: carrying out high-precision numerical integration by adopting an Adams-Bashform-Module solver; optimizing model parameters in combination with a local domain normalization pre-training strategy and a fusion loss function; abnormal detection is realized by calculating comparison between signal reconstruction deviation and a self-adaptive threshold value; a future anomaly probability is generated based on the trajectory prediction. According to the method, the problems of insufficient non-local dependence capture, poor time-varying interference robustness, high false positive rate and the like in the prior art are effectively solved, the accuracy and real-time performance of abnormal detection and prediction of biomedical signals such as electroencephalogram and electrocardiogram are remarkably improved, and the method is suitable for wearable medical equipment and clinical monitoring systems.

The invention discloses a continuous unsupervised domain adaptive fault diagnosis method. The method comprises the following steps: acquiring source domain and target domain fault data of rotating mechanical equipment; a prototype comparison learning strategy based on measurement and characterization learning is adopted, a fault diagnosis model is pre-trained through source domain data, a uniformly-divided unit hyperspherical feature space is obtained, and the fault diagnosis model comprises a feature extractor and a cosine similarity classifier; freezing a prototype obtained by source domain pre-training, and carrying out supervised training combined with source domain data on the fault diagnosis model to realize continuous domain adaptation to a target domain; source feature knowledge distillation and local domain alignment are carried out, mutual information loss between features and a prototype is maximized, and supervised training is carried out on a fault diagnosis model in combination with source domain data; and the fault diagnosis model is applied to actual fault diagnosis after completing learning of all current target domains. According to the method, new data distribution can be quickly and effectively adapted, the performance of the model on historical data distribution is kept, and knowledge forgetting is avoided.

The invention belongs to the technical field of computer man-machine interaction and automatic control, and particularly relates to an RPA mouse and keyboard control method and system for a Wayland desktop. The method comprises the following steps: S1, creating and registering a virtual input device supporting mouse and keyboard functions in a user space by accessing a virtual input device interface provided by an operating system kernel; s2, current activity display output is detected, and screen resolution information is obtained; s3, the server program monitors a preset local Unix domain socket path for receiving a control instruction from the client; s4, the client sends a control request to the local Unix domain socket path in a structured format, and the server receives and analyzes the control request; s5, the server side maps the analyzed control request into a corresponding kernel input event sequence according to the operation type; meanwhile, by writing event data into the virtual input device, an input event is injected into the system so as to simulate a control operation on the graphic desktop.

The invention discloses an intelligent diagnosis method and system for mechanical equipment faults under an open set. The method comprises the steps that a novel FCform encoder based on Fourier and convolution embedding is used for constructing a long-distance dependency relationship in a frequency domain, and more representative local domain invariant fault features are extracted; designing an open set dynamic sub-domain adaptive module based on DLMMD, and aligning condition feature distribution of known classes by covering potential unknown classes; in order to reduce the influence of empirical threshold setting on unknown class detection, decision boundaries of known classes and unknown classes are established by using an adaptive threshold learning method. And training a shared feature extractor and a fault classifier in the model through a minimum-maximum entropy game training strategy so as to suppress the problem of excessive confidence of the model possibly caused when the entropy value is minimized. According to the method, the problems of cross-domain knowledge migration of mechanical equipment faults and intelligent detection of unknown faults under the open set can be solved, and the method has a wide application prospect.

An improved computing system is arranged for cross-origin network communications on a single computing device. The system includes a processor, a networking module, and memory with software instructions arranged to operate a local computing server resource on a first local domain, instantiate a relay mechanism that has an iFrame and an invisible window, instantiate a local web server on a second local domain, install a service worker on the invisible window, receive a request for information at the local web server, verify a presence of the local computing server resource on the first local domain, communicatively connect the second local domain to the iFrame, and directly communicate, via the at least one networking module, at least one message between the local computing server resource on the first local domain and the local web server on the second local domain using the relay mechanism.

Systems and methods that allow a multihomed gateway to monitor its membership in individual domains in an Ethernet Virtual Private Network (EVPN) Data Center Interconnect (DCI) Network, and issue withdrawals from multihoming groups associated with those domains based on this membership, are disclosed. According to embodiments, links between a local domain and a gateway, and between the gateway and a remote domain, can be monitored at the gateway. When the link between the gateway and a domain goes down the gateway may withdraw itself from any multihoming group associated with the domain to which the link has been lost.

The invention relates to a general graph task method based on a large language model (LLM), which comprises the following steps of: firstly, serializing graph structure data into input suitable for the LLM, and injecting the structure information of a graph into a model through two codes, namely, RWPE (Random Walk Position Element) and Laplacian Position Element (LAPE) based on Laplacian decomposition to finish a general serialization process. Then, according to different tasks, different serializations are carried out, including subgraph segmentation and arrangement in a graph classification task and local domain sampling in a node classification task; after completely serialized data is obtained, knowledge injection is carried out in a mode of freezing and fine tuning a pre-trained large model, finally, the framework obtains excellent performance in the graph task field (graph classification and node classification), the framework effect is verified on multiple data sets, meanwhile, an interpretability experiment and a small sample scene verification experiment are carried out, and the method is suitable for large-scale application. The superiority of the frame is further proved.

The present invention discloses a three-dimensional magnetic resonance brain image analysis method and its application, and relates to the technical field of three-dimensional magnetic resonance brain image analysis. The present invention uses category information to perform feature alignment, and while adjusting the importance of marginal distribution and conditional distribution to domain adaptation, introduces MDD loss based on prototype representation to optimize the adversarial domain adaptation network loss; the cross entropy classification loss of the source domain, the attention consistency loss of the feature extraction network, the global domain adversarial loss for aligning the marginal distribution of the source domain and the target domain, the local domain adversarial loss for aligning the conditional distribution of the source domain and the target domain, and the MDD loss based on prototype representation together constitute the loss function of the network model. The present invention can not only achieve the simultaneous alignment of the marginal distribution and conditional distribution of the data, but also maximize the intra-class density, retaining and strengthening the corresponding extremely subtle intra-class and inter-class feature information.

A multiprocessor system includes a processor in a local domain that performs enumeration, and when a PCIe device in an external domain is not found by the processor in the local domain, an access conflict is avoided when PCIe switches in the two domains are interconnected, and the processor or a PCIe device in the local domain normally accesses the PCIe device in the external domain. Further, a communication path of the cross-domain access is established by using a bus and the PCIe switches in the two domains, and does not depend on forwarding of a processor.

This invention discloses a method and system for generating real-time content in a specific domain based on AI technology, belonging to the field of artificial intelligence technology. The method includes the following steps: 1. Constructing a local material knowledge base RAG_src in the form of word vector storage using AI technology, whose materials include multimodal materials such as text, voice, audio / video, or H5; 2. Constructing a local domain knowledge classification knowledge base RAG_policy in the form of file storage using AI technology, whose content includes domain specification documents and technical classification documents categorized by domain knowledge and information; 3. Constructing PageIndex by combining a Large Language Model (LLM) with RAG_policy; 4. Receiving content generation requests from real-time software, calling RAG_src and PageIndex through LLM to generate real-time content that meets the requirements of the corresponding scenario; 5. Checking the generated content and outputting compliant content. This invention solves the problems of non-standard content generation, disconnection from relevant standards, low reuse rate of proprietary materials, and low retrieval efficiency in existing technologies, achieving accurate, efficient, and real-time content generation. It is adaptable to various scenarios such as real-time content generation, content compliance checking, and real-time domain assistance in specific domains. Application scenarios include, but are not limited to, oral practice and oral examinations according to teaching syllabi.

The invention discloses a traction motor cross-equipment fault diagnosis method and system based on partial adversarial domain adaptive migration, and relates to the technical field of fault diagnosis. And a global domain discriminator model and a local domain discriminator model which can dynamically adjust loss weights. When fault data category spaces of the source domain motor and the target domain motor are inconsistent, the fault data category space of the source domain motor and the fault data category space of the target domain motor are not consistent. The transfer learning model can accurately select samples of fault categories related to a source domain and a target domain to avoid the problem of negative transfer caused by paying attention to unrelated outlier categories, effective transfer of fault diagnosis knowledge between the two domains under label offset is achieved, and the adaptability of the fault diagnosis method to cross-equipment fault diagnosis of an actual running motor is improved; the method is easy to implement, extra hardware equipment is not needed, and the real-time operation health monitoring level of the motor and the train can be improved.

The invention relates to the technical field of data query, in particular to a bank data query intelligent question and answer interaction system based on a large model. Comprising a question and answer interaction end, a bank database, a data retrieval module, a local domain extraction module, a cross-domain extraction module and a reply integration module, in order to solve the technical problems of tedious cross-domain extraction, dependence on manual authorization and limited question and answer range caused by data dispersion, the invention constructs a large-scale model driving system fusing natural language understanding, intelligent permission routing and collaborative workflow. Inherent data islands in banks are converted into logically associated data networks and efficient intelligent collaboration, so that non-technical personnel can accurately obtain cross-system complex business data in a dialogue mode, automation and refinement of data security management and control and cross-department collaboration process reconstruction are realized at a bottom layer, and the data security management and control efficiency is improved. Therefore, on the premise that financial compliance and data security requirements are strictly met, fundamental improvement of data query efficiency, depth and a cooperation mode is realized.

The invention discloses a redundancy optimization-based power grid data consistency method, which comprises the following steps of: dividing a whole network into a plurality of local domains by utilizing reputation data and topological information of nodes, electing a global committee, further performing dynamic influence evaluation on a transaction request based on a network domain division graph, and according to an evaluation result, determining the consistency of the transaction request. High-frequency local transactions are limited in a local consensus domain for closed-loop processing, and only transactions related to global influences are submitted to a global committee. Through the mode, the communication redundancy of the whole network is remarkably reduced, so that the concurrent processing capability and expansibility of the system are greatly improved while the data consistency is guaranteed.

This invention discloses a generalized identification method and system for slope rock mass structural surfaces based on multi-level domain adaptation. This invention relates to the field of slope rock mass identification technology. By constructing four levels—global domain adaptation, regional domain adaptation, local domain adaptation, and sample-level adaptive adaptation—this invention achieves hierarchical and progressive evolution of model capabilities. A variational autoencoder is used to explicitly decouple features into domain-invariant and domain-specific parts, enabling the model to distinguish between the essential geometric features of rock mass structural surfaces and the weathering and unloading features of specific regions. Through regional meta-learning training, it can quickly adapt to new regions with only a small number of labeled samples.

The invention discloses a network access method and device in a DNS hijacked scene and a related product, and relates to the technical field of network security. The method comprises the following steps: firstly, checking a local domain name resolution cache of a client, preferentially multiplexing a valid IP address cache, if the valid IP address cache is invalid, acquiring an IP address list corresponding to a target domain name through an IP direct connection trusted service interface, and bypassing a DNS resolution link; an IP address set with timeliness is generated and stored persistently, and address validity is guaranteed through regular refreshing of a double-judgment mechanism; a main domain name, a standby domain name and a local cache IP address are obtained based on the set, and a domain name-first progressive failover access strategy is generated in combination with multi-network configuration; and when the access of the main domain name and the standby domain name fails, automatically switching to the local cache IP address to initiate the access. According to the method and the device, stable network access in a DNS hijacking scene is realized, and reliable network access guarantee is provided for various software which depends on domain name access.

The present application relates to the technical field of artificial intelligence, in particular to a domain self-adaptive large model fine-tuning method and system based on enterprise private data, the method comprising: each participating client extracts the feature representation of local data using a unified pre-training model, and generates a prototype set representing the category features of the local domain by using mean, clustering or random sampling methods; uploading the prototype set to the server after adding Gaussian random noise to realize differential privacy protection; the server aggregates all client prototypes to construct a global prototype training set, and performs global training of the adapter module in a single communication round using supervised learning classification loss without averaging the same prototypes of different clients, obtains a global adapter model and distributes it to the client; the client initializes the local model based on the global adapter model, fine-tunes it using a small amount of labeled data, and uses the global model as a teacher to prevent catastrophic forgetting through knowledge distillation, achieving a balance between global knowledge retention and local performance improvement. The present application can solve the problem that the existing technology cannot effectively align the domain features of different enterprises, resulting in a significant decline in the performance of the model in cross-domain reasoning tasks.

This application provides a data security sharing method, apparatus, and system based on a usable but invisible mechanism, applied in the field of cross-domain data sharing security technology. It constructs a secure sharing mechanism based on a three-dimensional collaborative approach of "spatiotemporal-object-rights," deeply integrating a spatiotemporal grid coding system with a digital object identification system to generate unique identifiers with spatiotemporal semantics. This is combined with a rights root mechanism to complete data ownership registration. During data circulation, relying on data network infrastructure and digital contract mechanisms, usage requests are routed to the data provider side, and a trusted execution environment bound to the contract is scheduled. This ensures that the original data completes secure computation without leaving the local domain. Simultaneously, traceable evidence information and measurable encrypted credentials are automatically generated along with the computation results. Ultimately, this achieves trusted sharing throughout the entire lifecycle of data—data that remains within the domain, usable but invisible, controllable, and measurable—providing fundamental support for a data-driven society.

A UMR auto-discovery mechanism which allows a gateway EVPN router to advertise a UMR route and to suppress redistribution of remote domain MAC-IP routes. In one embodiment, upon joining a local EVPN domain of a network, a network device (an EVPN router) advertises its UMR capability, which may include UMR installer capability or UMR originator capability. The UMR-capable devices in the network are also advertised to the newly added network device. A UMR originator (e.g., gateway) with knowledge of the UMR capabilities of the devices in the network then generates and sends a route list to the devices, where the route list includes a UMR route, or MAC / IP routes, or both, depending upon whether all of the network devices are all UMR-capable, none are UMR-capable, or the devices include both UMR-capable and non-UMR-capable devices, respectively.

The invention relates to an electroencephalogram self-adaptive classification method based on a dynamic united domain adaptation network and application. The method comprises the steps that firstly, EEG signals are collected and preprocessed and then transmitted to an upper computer; after receiving the preprocessed EEG signal, the upper computer performs signal feature extraction and classification based on a dynamic joint domain adaptation network DJDAN model, and the DJDAN model comprises a feature extractor, a classifier, a global domain discriminator and a local domain discriminator; the upper computer converts the motion intention label output by the classifier into a control signal required by rehabilitation equipment; the rehabilitation equipment generates feedback signals according to the training actions of the patient and transmits the feedback signals to the upper computer, and the upper computer records and evaluates the training progress of the patient. The motion intention recognition based on the electroencephalogram signals can be realized, the method is applied to an intelligent rehabilitation training system of a stroke patient, and the rehabilitation effect and the motion ability of the patient are improved.

The present application relates to the technical field of nondestructive testing of underground engineering, ground penetrating radar image processing and deep learning target detection, and specifically discloses a method for detecting internal defects of a water conveying tunnel, wherein a source domain data set is constructed by generating simulated ground penetrating radar data with defect labels through time domain finite difference forward simulation simulation, measured ground penetrating radar data is collected in the tunnel and preprocessed to construct a target domain data set, a YOLO target detection network comprising a network backbone, a feature fusion network and a detection head is built, then the source domain data set is used for supervised training, the source domain data set and the target domain data set are jointly input for global domain adaptive training, followed by local domain adaptive training, and finally the measured ground penetrating radar data to be detected is input into the network to output the defect detection result.

The invention discloses an intelligent comparative analysis method and system for a network security situation, and relates to the technical field of network security, and the method comprises the steps: constructing a multi-source coupling index E and a disturbance anomaly index Ycz on the basis of obtaining a standard environment perception data set and a standard communication perception data set; quantitative judgment on whether environmental disturbance can reasonably explain communication abnormity or not is achieved. When the disturbance anomaly index Ycz is higher than a set threshold value, a node transverse behavior structure analysis mechanism is further triggered, the sensing node is divided into a local domain Dm based on physical adjacency and topological adjacency, and a behavior structure offset function Mm is calculated based on the disturbance anomaly index of each node in the local domain. According to the structural analysis mode, abnormal outliers, synchronous abnormal groups or potential attack source nodes in a local structure can be effectively recognized, the linkage recognition capability of a system on local explosive attacks or node out-of-control phenomena is improved, and comprehensive perception upgrading from single-point behavior abnormity to structure-level situation migration is achieved.

The application discloses a multi-working-condition industrial process fault diagnosis method based on a dynamic adaptive domain adversarial network, and comprises the following steps: standardizing process data and taking a sliding window to obtain dynamic data; training a dynamic adaptive domain adversarial network by using labeled dynamic data of a source working condition and unlabeled dynamic data of a target working condition; the dynamic adaptive domain adversarial network comprises a feature extractor, a label classifier, a global domain discriminator and a local domain discriminator; the global domain discriminator aligns the marginal distribution of the source working condition and the target working condition; the local domain discriminator aligns the conditional distribution of the source working condition and the target working condition; a kind of learnable parameter is proposed to adaptively evaluate the relative importance of the marginal distribution and the conditional distribution, so as to better extract domain-invariant features; a kind of common center loss is proposed to improve the inter-class separability and intra-class compactness of data, and further improve the fault diagnosis precision; and the label classifier is used to predict the fault category of a sample. The application establishes an accurate fault diagnosis model for the target working condition.

The present invention belongs to the field of machine learning privacy and security technology, and provides a federated learning hybrid aggregation mechanism under a blockchain architecture. Using local servers as management and miner nodes, the network is divided into multiple local domains consisting of local servers and terminal devices, and an inter-domain network consisting of local server connections. In the local domain, a coding algorithm that satisfies local differential privacy is used to select gradients to defend against gradient exponential attacks; a perturbation algorithm that satisfies local differential privacy is used to perturb gradients to defend against gradient content attacks; the public key of the local server is used to encrypt the gradient to defend against eavesdropping attacks; a shuffle server is used to shuffle and sample user updates to defend against user identity attacks. A local server is used to decrypt and aggregate updates within the domain, and an update evaluation and consensus mechanism based on Shapley values ​​is used between domains to issue rewards and update models. Users within the domain share the rewards equally to ensure user privacy.

The invention discloses an open set domain adaptive blast furnace fault diagnosis method based on reciprocity point learning, and belongs to the field of industrial process monitoring and intelligent diagnosis. According to the method, an open set domain adaptation framework is provided for the problems of fault sample scarcity, data distribution drift and unknown fault category occurrence in the blast furnace production process. The method comprises the following steps: firstly, constructing a mixed CNN-ViT feature extractor, and fusing local and global features to capture the slow time-varying feature of a blast furnace fault; and secondly, designing a prototype-reciprocity double classifier, constraining the open space risk through reciprocity point learning, and realizing effective distinguishing of known and unknown faults. And an entropy-based adaptive threshold mechanism is further introduced to dynamically identify an unknown fault. And finally, domain invariant feature learning and cross-working-condition knowledge migration of known faults are realized through local domain adaptation and joint distribution alignment. The method has excellent fault diagnosis precision and robustness in multi-working-condition and multi-source-domain data of the blast furnace, and the recognition capability of unknown faults is remarkably improved.

The invention relates to a dynamically-configured VPN server multi-domain name shunting method and device, computer equipment and a medium, and the method comprises the steps: receiving the flow of a VPN client, and recognizing whether the flow belongs to DNS query flow or non-DNS query flow; when the flow belongs to DNS query flow, forwarding the flow to a local DNS query server, and when DNS response information returned by the DNS query server is obtained, extracting an IP address and domain name information from the DNS response information, and updating the IP address and domain name information to a local IP domain name mapping table; when the traffic belongs to non-DNS query traffic, performing domain name resolution; when the analysis is successful, executing a domain name shunting operation according to the domain name information obtained by analysis, and updating the target IP address of the traffic and the domain name information obtained by analysis to a local IP domain name mapping table; and when the analysis fails, querying a local IP domain name mapping table according to the target IP address of the flow, and executing domain name shunting operation according to the queried domain name information. According to the method, domain name shunting can be accurately and safely carried out on the encrypted traffic in real time.

The invention belongs to the technical field of computer resource sharing, particularly relates to a collaborative system, method and equipment for realizing cross-domain signal source pushing and keyboard and mouse sharing, and aims to solve the problem that signal source pushing and KVM sharing cannot be performed in a cross-domain manner. The system comprises a business module which is used for an interaction terminal of a local domain to obtain any signal source and push the signal source to a target interaction terminal of the local domain or a target interaction terminal of an opposite-end domain; the resource module is used for configuring and managing resource permissions including a keyboard and a mouse of the domain according to the target interaction terminal of the domain; or the resource permission configuration module is used for configuring resource permissions, including the keyboard and the mouse, of the local domain to the cross-domain module of the local domain according to the target interaction terminal of the opposite-end domain; and the cross-domain module is used for realizing cross-domain sharing of resource permissions including a keyboard and a mouse between the interaction terminal of the local domain and the target interaction terminal of the opposite-end domain. According to the invention, cross-domain signal source pushing and keyboard and mouse sharing of the system are effectively realized.