164 results about "Inter-domain" patented technology

The invention provides a cross-domain equipment fault diagnosis method and system based on large and small model cooperation, and relates to the technical field of equipment fault diagnosis. According to the method, the causal field generalization structure is introduced into the small model, explicit decomposition is carried out on the stable causal law and the field specific difference, and meanwhile, the causal field generalization structure is corrected by using the large model, so that the small model can automatically identify and retain the causal relationship which is universally applicable to each device and each field; therefore, the influence of inter-domain distribution difference is effectively eliminated. Theoretical analysis shows that the generalization error of the model mainly depends on the accuracy of the stable causal item, and the structure can minimize error drift caused by distribution drift. Therefore, the robustness of health state evaluation and fault prediction can be remarkably improved in a cross-domain scene, and the fault diagnosis model can still keep the prediction capability close to the training domain level under the condition of no target domain annotation data.

The invention discloses a wind power prediction method based on multi-source domain deep transfer learning, and relates to the field of new energy power prediction.The method comprises the steps that the data distribution difference between a multi-source domain and a target domain is reduced through an Euclidean alignment method, and the maximum mean value difference between the domains after optimization is obtained; setting a migration weight factor for each source domain based on the maximum mean value difference, and constructing a weighted migration training set; constructing a multi-source domain deep migration learning model, and carrying out migration training on the model based on a weighted migration training set; and finely tuning the pre-training model according to a small amount of data of the target domain to obtain a target domain wind power prediction model. The wind power data distribution difference between the multi-source domain and the target domain is reduced through the Euclidean alignment method, the knowledge migration efficiency is improved, and the negative migration risk is reduced; setting a migration weight factor for each source domain based on inter-domain MMD, so that the model preferentially learns high-correlation source domain wind power characteristics; and the multi-source domain deep migration learning model is fused into a dynamic weight module, so that low-efficiency migration is avoided.

The invention belongs to the technical field of adversarial samples. The invention provides a robust generalization-oriented few-sample continuous confrontation defense method. According to the embodiment of the invention, edge distance loss is resisted, in the pre-training stage, by maximizing the distance between the clean sample and the model decision boundary, the discrimination capability of the model for easily confused samples near the boundary is explicitly improved, the robustness of the model for the subsequent few-sample adversarial adaptation stage is enhanced, and the generalization capability of the clean sample and the adversarial sample is improved. Gaussian mixture model prototype playback is also provided, modeling is carried out on historical adversarial domain feature distribution by using the Gaussian mixture model, pseudo features are generated for knowledge playback, original adversarial samples do not need to be stored, and the robustness of the model in new and old adversarial domains is improved. Besides, by designing multi-domain balance loss, in multi-domain continuous adversarial training, updating is facilitated for most historical domains by constraining a model updating direction, inter-domain loss variance is reduced, and cross-domain balance optimization is realized.

The invention discloses a data transmission method and device between security domains, a storage medium and electronic equipment. Relates to the field of data transmission, and the method comprises: in a front-end application of a first security domain, in response to an interaction request triggered by a user, generating a request message containing a unique identifier, and issuing the request message to a message queue service in the first security domain, the message queue service being used for a large model processing service in a second security domain, consuming the request message according to a pre-configured security access strategy, and writing response data of the large model into a shared cache service of a first security domain; and initiating a query request to the shared cache service to obtain response data corresponding to the interaction request from the shared cache service, and displaying the response data to the user through a user interface of the front-end application. The problem that data transmission efficiency is low when data transmission between security domains in a one-way network isolation environment is realized by depending on a manual or semi-automatic off-line ferry mode in the prior art is solved.

The invention belongs to the technical field of pipeline safety assessment, and discloses a multi-dimensional information interconnection and autonomous evolution collaborative pipeline full-state safety assessment method. And capturing a high-order relationship of data through double hypergraph reasoning of the instance-level hypergraph and the modal-level hypergraph to realize efficient interconnection. According to the method, mode-level and instance-level hypergraph information features are extracted through hypergraph information propagation, high-order correlation is mined through double-graph information aggregation, cross-mode and cross-instance consistency information and exclusive information are output after feature recombination, multi-dimensional data deep fusion is promoted, and high-quality data support is provided for follow-up pipeline full-state safety assessment. A two-stage autonomous evolution mechanism of intra-class progressive calibration and inter-class knowledge migration is respectively adapted to slight fluctuation and significant change scenes of the deep sea environment: precise adaptation of environment perturbation is realized through dual-branch feature extraction and dynamic weight adjustment in a domain; model parameter dynamic optimization is completed between domains through spatial-temporal feature clustering and cross-domain knowledge migration, and dynamic environment self-adaption can be achieved without manual intervention.

The invention relates to the technical field of network communication, and provides a hierarchical reinforcement learning scheduling and routing method for a multi-domain TSN, and the technical scheme comprises the steps: collecting the global state information of the multi-domain TSN, and carrying out the coding processing of the global state information to generate comprehensive state representation; performing a cross-domain routing decision, and outputting an inter-domain path and a time delay budget of a cross-domain flow; executing intra-domain scheduling, determining a sending sequence and a specific path of a domain flow, and generating gating list configuration; through hierarchical coordination and strategy optimization, a cross-domain routing decision and intra-domain scheduling are updated based on reward feedback of an intra-domain scheduling result; and deploying the finally updated intra-domain scheduling to a switch of the multi-domain TSN network for execution, thereby realizing deterministic transmission of the time-triggered flow. According to the invention, through a layered agent architecture, a hybrid neural network coding mechanism and a cross-domain collaborative optimization strategy, challenges of complexity, expandability, dynamic adaptability and the like of a joint routing and scheduling problem in a multi-domain TSN environment are effectively solved.

The invention provides a multi-source domain invariant acoustic feature extraction method and system for an equipment operation state, and belongs to the technical field of equipment maintenance, and the method comprises the steps: constructing a multi-source domain invariant acoustic feature extraction network based on a DANN model, and the network comprises a feature extractor, a classifier, a domain discriminator and a multi-domain acoustic feature class boundary constraint module; the feature extractor extracts high-dimensional features of sound signals, the classifier carries out fault mode recognition, and the domain discriminator realizes domain prediction. The multi-domain constraint module generates an embedding space, and calculates the maximum mean value difference, the local maximum mean value difference and the Euclidean distance among different source domain features. The network constructs a loss function by taking minimization of inter-domain difference and classification loss and maximization of inter-domain distance and domain discrimination loss as targets, and carries out adversarial training through multi-source tagged acoustic data. According to the method, domain invariant features are extracted by using adversarial learning and hidden space constraint alignment of multi-source domain acoustic features, so that the influence of feature offset under a cross-working-condition condition is effectively reduced, and the fault mode recognition accuracy is improved.

The invention relates to the field of information security, and discloses a cross-domain quantum key charging method and system.The method comprises the steps that a first key charging site responds to a first service terminal request, analyzes a target terminal identifier and reports step by step when cross-domain judgment is conducted; the cross-domain key charging management center generates a cross-domain master key based on a quantum true random number according to the target attribution, and distributes the cross-domain master key to a source domain key charging service center and a target domain key charging service center through a quantum secure channel; the service centers of the two parties negotiate a derived session key by using the shared master key and issue the derived session key; and after receiving the session key, the site generates a service key in combination with the service context and charges the service key to the terminal. The system also integrates a dynamic request triggering mechanism based on consumption rate, a hybrid encryption channel mechanism and a storage unit physical overwriting destruction mechanism. According to the method, a hierarchical cascade architecture is adopted, and a quantum entropy source and a post-quantum algorithm are combined, so that the problem of trust transfer between heterogeneous domains is solved, and quantum computing attack-resistant key full-life-cycle safe closed-loop management is realized.

The invention relates to the technical field of video analysis and processing, can be applied to the field of science and technology finance / digital medical treatment, and discloses a video key frame extraction method based on cross-domain attention and related equipment. The method comprises the following steps: respectively carrying out size adjustment and pixel normalization on a source domain video and a target domain video to obtain corresponding video frames; inputting the source domain frame into a pre-training visual feature extraction model, inputting the target domain frame into a convolution feature extraction model, and respectively extracting frame features; calculating the frame-level semantic correlation of the two domain features based on a cross-domain attention mechanism, and generating a cross-domain attention weight matrix; weighting the frame features of the target domain according to the weights to obtain spatial-temporal feature representation with semantic alignment; generating a key frame selection mask according to inter-domain feature differences, and determining a key frame set in combination with sparsity constraint and time sequence coherence constraint; and removing redundant frames based on inter-frame similarity, and performing time sequence smoothing processing to obtain a final key frame sequence. According to the method, cross-domain frame alignment is more accurate, and dynamic frame screening is more stable.

The invention belongs to the technical field of security authentication, and discloses a distributed energy storage equipment cross-domain security access authentication method and a related device, attribute information is submitted after zero-knowledge proof verification, and a certificate issuing party is triggered to generate a verifiable certificate based on a signature algorithm; then generating a zero-knowledge proof-based randomized verifiable representation based on the verifiable credentials; and finally, the energy storage device sends the generated randomized verifiable representation to a verifier to complete access authentication. According to the invention, cross-domain authentication is realized through a distributed architecture and verifiable certificates without depending on a single centralized mechanism. Verifiable certificates adopt a uniform format and support mutual recognition among different trust domains. And the certificate cannot be associated with a specific equipment identity after randomization processing, so that anonymous authentication is realized. And the verification process is realized through zero-knowledge proof, and the equipment can prove the identity legality without leaking sensitive information, so that the authentication security and privacy are effectively guaranteed.

The invention discloses a multi-label field adaptive method based on prompt driving, which belongs to the technical field of computer vision and comprises the following steps: generating multi-dimensional semantic text description containing visual attributes, semantic levels and category names of common scenes for each type; embedding the semantic text description into an optimizable category vector, and embedding the optimized category vector into a CLIP text prompt; extracting and projecting multi-layer style statistical characteristics of the image, and synchronously realizing image-text cross-modal alignment and source-target domain distribution alignment in a frame through a cross-domain style mapping network and various alignment losses; semantic priori knowledge of CLIP is combined with a label co-occurrence mode of a source domain, so that the model can sense and adapt to a possibly changed label dependency relationship in a target domain; performing semantic propagation on label embedding by using a graph convolutional network; the whole system is jointly optimized through a multi-task loss function, and end-to-end cross-domain multi-label classification is achieved.

The invention discloses a smart city power marketing service simulation evaluation method and system based on digital twinning. The method comprises the steps of obtaining power marketing service data, performing space-time semantic analysis, extracting attribute dimensions, establishing a cross-domain attribute association graph, establishing a cross-domain attribute association graph, and constructing a digital twinborn model; the target marketing service process is decomposed into service units, the service units are decomposed, a local simulation domain is constructed for parallel simulation, and a combined simulation result is obtained through an inter-domain coupling operator; and comparing a simulation result with real data, identifying virtual and real deviations, performing reverse traceability decomposition on virtual and real deviation distribution, mapping the virtual and real deviation distribution back to each local simulation domain, determining a local contribution magnitude of each local simulation domain, and constructing a hierarchical evaluation index to obtain a service quality evaluation result. According to the invention, accurate evaluation of the power marketing service is realized, and the service quality supervision efficiency is improved.

This invention discloses a method and system for cross-domain fault diagnosis of rolling bearings with unknown inter-domain data label relationships. It relates to the field of rolling bearing fault diagnosis technology and aims to solve the problem of low fault diagnosis accuracy in existing transfer learning methods due to unknown data label relationships between the source and target domains. The key technical points of this invention include: constructing a comparative general domain adaptation model; introducing a BYOL network to mine the unique structure of the target domain data and proposing an entropy separation strategy to reject unknown class samples; simultaneously, designing a source class weighting mechanism to improve the transfer semantic enhancement method, assigning different class-level weights to the source classes, so that the feature distributions of the two domains are better aligned in the shared label space, further constructing a fault diagnosis model; and performing fault diagnosis through the trained fault diagnosis model. This invention exhibits superior fault diagnosis accuracy under different operating conditions.

The application discloses a kind of based on two-dimensional coupling chaos vehicle CAN bus lightweight security communication method, belong to vehicle network security communication technical field.This method is applied to the layered vehicle network including gateway electronic control unit GECU, domain control unit DCU and electronic control unit ECU, including four stages of system initialization, identity authentication, key distribution and secure communication.Through two-dimensional coupling chaotic mapping, random challenge value, authentication random quantity and chaotic seed are generated, and combined with global round parameter epoch and domain key evolution parameter rand, a layered key management mechanism is constructed to realize the local derivation of inter-domain and intra-domain shared session keys.The sending node uses the shared session key to perform lightweight encryption and authentication on the service data, and encapsulates the ciphertext and authentication tag in the same CAN message for transmission.This method improves the communication performance of vehicle CAN bus without significantly increasing communication overhead and computational overhead.

The invention discloses a crimping quality diagnosis method and system based on feature guidance and domain adversarial migration, and belongs to the technical field of electrical system quality detection. According to the method, an integrated diagnosis system is constructed, and multi-domain feature extraction and preprocessing are carried out; a single-domain hybrid guide network is adopted to realize high-precision classification; a cross-domain dual-alignment domain adaptive network is adopted to realize discriminant alignment and statistical alignment of inter-domain features, domain differences are reduced, and multiple input modes are adapted; and finally, generating a high-precision and explainable comprehensive quality diagnosis result. According to the method, comprehensive coverage of crimping quality from single-domain high-precision detection to cross-domain self-adaptive diagnosis is realized, and compared with a traditional crimping quality detection method, the technical scheme of the invention significantly improves diagnosis precision, cross-domain generalization ability and robustness to category missing scenes, and improves the accuracy of crimping quality detection. Therefore, an intelligent and automatic innovative solution is provided for the quality control of the crimping process of a modern electrical system.

The invention relates to the technical field of rotating machinery state monitoring and fault diagnosis, and discloses a rotor bearing fault mode identification method based on a physical sample and a domain alignment network, so as to solve the problem of insufficient marked samples in a model training process. According to the method, firstly, a lumped parameter model is established, a complete defect sample is generated, and then simulation data performance is optimized by using a Pearson correlation coefficient (PCC); then, an enhanced joint distribution adaptation (EJDA) mechanism is adopted to coordinate inter-domain distribution, overall distribution and inter-class distribution in the domain are effectively adjusted, and adaptive domain confusion is achieved; and finally, in order to ensure robust feature classification and extraction, a self-adaptive softmax (A-softmax) loss function is introduced to improve the classification performance of a cross-domain diagnosis task.

The invention relates to the technical field of network information security, and particularly discloses a multi-domain network attack-oriented unsupervised integrated detection method and system. In the scheme, a multi-channel automatic encoder of an integrated network can design a differentiated structure according to feature dimensions and distribution rules of different attack domain data; the multi-channel design can accurately extract key features of data of each domain, representation deviation caused by inter-domain feature distribution conflicts is avoided, hidden space representation better fits essential features of attacks of each domain, and robustness is improved. The channel automatic encoder can realize effective information focusing through parallel dimension reduction and feature fusion: on one hand, each channel can adopt different dimension reduction paths to mine effective information from high-dimensional data in a multi-view manner, and on the other hand, the integrated network performs fusion processing on multi-channel dimension reduction results, local noise of each channel can be filtered, and the multi-channel dimension reduction result is obtained. The optimized hidden space representation has the characteristics of high information density and low noise interference at the same time, and the robustness and the detection precision can be improved.

The present disclosure relates to an electronic device and method in a wireless communication system, and a computer-readable storage medium. Provided is an electronic device for a spectrum management apparatus in a spectrum-sharing system. The spectrum-sharing system maintains a single blockchain and is partitioned into a plurality of spectrum-sharing autonomous domains; each spectrum-sharing autonomous domain comprises a corresponding spectrum management apparatus and a plurality of base stations; and the spectrum management apparatuses in the plurality of spectrum-sharing autonomous domains realize inter-domain communication by means of a satellite. The electronic device comprises: a processor; and a memory comprising a computer program code, wherein the computer program code, when executed by the processor, causes the electronic device to perform operations comprising: selecting a spectrum-sharing task; by means of an intra-domain task processing and consensus process, generating an autonomous domain candidate block associated with the spectrum-sharing task; confirming, via a satellite, the autonomous domain candidate block by means of an inter-domain task processing and consensus process; and adding the confirmed autonomous domain candidate block as a block of a blockchain.

The application provides an aero-engine partial unsupervised cross-domain fault diagnosis method based on an inter-domain attention mechanism, designs an inter-domain attention encoding module, and gives different weights to source samples belonging to an outlier health space and source samples belonging to a shared health space through the inter-domain attention mechanism, so as to reduce the negative transfer caused by the outlier source samples in the domain alignment process. And through the adversarial network, the source samples belonging to the shared health space and the target domain samples are aligned to realize positive transfer. Experiments prove that the application is superior to other similar methods in the aero-engine partial unsupervised cross-domain fault diagnosis task. The method is suitable for real industrial application scenarios, and application of the method can assist researchers to fully utilize a large amount of historical accumulated data to establish a high-accuracy fault diagnosis model suitable for each monitored engine, which has important significance for engine state monitoring and maintenance decision.

The application discloses an edge computing network architecture providing deterministic quality of service, which mainly comprises a service provider, a user terminal, a base station, a router and an edge computing node, wherein the service provider receives a task of the user terminal and feeds back relevant configuration parameters for processing the task to the user terminal and relevant equipment on a task transmission / processing path, the base station receives the task of the user terminal and forwards the task to the router, and the router receives the task sent by the base station and forwards the task to the edge computing node; a three-part domain deployment mode is adopted, including three independent domains of a wireless domain, a wired transmission domain and a computing domain, and an inter-domain traffic shaping mechanism is adopted. The method transplants the deterministic network technology into the edge computing network, solves the problem that the traditional deterministic network technology cannot be adapted to a heterogeneous medium scene, and can provide deterministic quality of service in a dynamic and unpredictable environment.

Various methods and processes, apparatuses or systems, and media for performing dynamic observations of inter-domain event processing in real-time are disclosed. The present disclosure provides receiving an inter-domain event; transmitting, to an event streaming platform, the inter-domain event as a data stream; extracting a topic for the serialized inter-domain event; and identifying, at an observer application, a workflow corresponding to the identified topic. The present disclosure further provides applying an event template corresponding to the deserialized inter-domain event, and based on the event template applied, dynamically applying a corresponding rule set without first requiring storage of the inter-domain event as a data object for capturing attributes corresponding to the inter-domain event on the fly.

A centerless adaptive distributed joint simulation scheduling system comprises a plurality of local scheduling domains, data communication among the local scheduling domains is realized through'inter-domain interaction interfaces', each local scheduling domain comprises a local master node and a plurality of local slave nodes forming a subsystem, the local master node communicates with the local slave nodes, and the local slave nodes communicate with the local master node. The method is used for automatically carrying out local master and slave node dynamic division, intelligent model distribution based on calculation complexity, time sequence scheduling of self-adaptive propulsion step length, step length optimization and data interaction based on the least common multiple, and subsequent simulation data prediction based on a fault breakpoint.

The invention discloses a remote sensing image unsupervised domain adaptive semantic segmentation method and system fusing frequency domain and spatial domain features, and aims to solve the problems of insufficient cross-domain feature alignment and limited segmentation precision caused by video domain information negligence and single domain discriminator scale in the prior art. According to the method, firstly, a source domain data set and a target domain data set are obtained, and then an unsupervised domain self-adaptive framework is built. A generator of the framework extracts multi-scale features through a visual backbone network, frequency domain and spatial domain information is compensated through a bidirectional recursion fusion module, then the features are processed by a content perception decoder containing four decoding stages, and each decoding stage is configured with a weight fusion module and an efficient content perception module to optimize feature representation; the discriminator distinguishes the attribution of the feature domain through four cascade stages, and each cascade stage is matched with a down-sampling unit and an efficient content awareness module to strengthen domain attribute feature extraction. Inputting a source domain training set and a target domain training set into a framework for training, back-propagating optimization parameters by using a multi-joint loss function, and storing an optimal model with the highest average intersection-union ratio in the training process; and finally, processing a target domain remote sensing image by using the model to realize semantic segmentation. According to the method, the feature distribution difference between the source domain and the target domain is reduced through deep fusion of the frequency domain and spatial domain features, and the problem of low segmentation precision caused by inter-domain difference of an existing unsupervised domain adaptive semantic segmentation method is effectively solved.

The invention relates to multi-domain virtual network security mapping for cross matching and optical path separation fault tolerance, and belongs to the technical field of optical fiber communication. According to the method, a double-layer block chain network architecture is adopted, a main chain is used for inter-domain resource collaborative scheduling and information synchronization of an optical network, and a secondary chain realizes intra-domain resource autonomous management and trust maintenance of the optical network; in the virtual node mapping stage, mapping preference values between virtual nodes and physical nodes are introduced to construct a mapping bigraph, and virtual node mapping is optimized through a cross matching mechanism; in the virtual link mapping stage, the security and reliability of quantum key distribution are guaranteed by adopting multi-path routing and a security redundant path, and an encrypted data transmission optical path based on minimum spectrum consumption is adopted; punishing malicious nodes according to a virtual network mapping result, and tracing and updating credibility values of non-malicious nodes participating in virtual network mapping through block information. According to the method, the security and the resource utilization rate of virtual network mapping in the optical network multi-domain environment can be improved.

The invention provides a high-throughput random number generation method and generator based on a quantum entanglement source, and relates to the field of quantum computation.The high-throughput random number generation method comprises the following core steps that firstly, in the entanglement debiasing process, a quantum sequence entangled with a random seed is constructed through inference, complete extension mapping of an anti-correlation entanglement quantum sequence is achieved, and the quantum sequence is generated; and generating a random bit sequence with doubled length and unbiased statistics. Secondly, in the full-permutation expansion process, the inter-domain correlation of the full-permutation sequence is eliminated through a shuffling method while sequence length exponential growth is achieved; and thirdly, in the layered asynchronous arrangement process, the intra-domain correlation is further eliminated through two-layer segmentation, asynchronous arrangement and shuffling processing, and finally a random bit sequence which meets the statistical quality requirement and has the length. According to the method, through removal of intra-domain and inter-domain correlation, the unpredictability of output random numbers is remarkably enhanced; and exponential-level high-throughput output is realized while the output random number is ensured to pass through all test items of the NIST SP 800-22.

This invention discloses a domain-specific intelligent routing method for large-scale low-Earth orbit (LEO) satellite networks, belonging to the field of satellite communication technology. This method constructs a network model based on multi-dimensional resource characteristics and combines an improved Transformer architecture regional traffic prediction algorithm to provide resource constraints and spatiotemporal correlation characteristics for routing decisions. Based on geographical region division into virtual nodes and autonomous domains, the LEO satellite network routing process is decoupled into intra-domain and inter-domain parts. Intra-domain routing employs a multi-agent deep Q-network, while inter-domain routing uses a summation tree-based deep Q-network. Routing and resource scheduling algorithms are designed separately for each domain. Through hierarchical routing and decoupled resource scheduling, a domain-specific intelligent routing system with both flexibility and scalability is constructed. This invention significantly enhances the service capabilities of LEO satellite networks under extreme scenarios such as high-frequency topology changes, strictly limited resources, and sudden spatiotemporal changes in traffic, providing a highly reliable, low-latency routing solution for integrated space-ground communication, emergency disaster early warning, and global real-time data transmission.

A collaborative diagnostic method and system for equipment faults based on secure multi-party computation is disclosed, relating to data processing systems or methods applicable to administrative, commercial, financial, management, supervisory, or predictive purposes. The method includes: acquiring equipment operating data within each security domain; generating fault feature summaries by grouping data according to equipment type and parameter ranges; calculating multidimensional similarity of fault features between domains based on equipment type consistency, overlap of state feature ranges, and time delay; constructing a cross-domain fault propagation network; identifying fault loop propagation paths and fault convergence nodes; and finally outputting the analysis results. Implementing this method can improve the accuracy of identifying cross-domain fault propagation paths and key nodes without sharing raw data details across security domains.

The invention discloses an insulator damage detection model training method and system based on domain self-adaption, and belongs to the field of computer vision, and the method comprises the steps: obtaining a labeled source domain image and an unlabeled target domain image, inputting an original model for training, carrying out the convolution of the two types of images, extracting initial features, carrying out the multi-round self-attention enhancement, and carrying out the training of the original model. In each round, dominant features of inter-domain soft alignment are generated through cross attention; screening features according to round intervals, and calculating entropy values of target domain and source domain features and dominant features thereof to obtain entropy loss; calculating an adversarial loss based on the cross entropy loss of the features after the first round of self-attention enhancement and the focus loss of the last round; calculating detection loss based on the source domain dominant features of the last round; and performing weighted fusion on the three losses to generate total loss, and updating model parameters. Therefore, by implementing the method, the problem that the detection performance of an insulator damage detection model is remarkably reduced due to domain migration under a cross-environment condition in the prior art can be solved.

A machine implemented method includes sensing entities in first and second domains. If a first stimulus is present and an entity is in the first domain, the entity is transferred from first to second domain via a bridge. If a second stimulus is present and an entity is in the second domain, the entity is transferred from second first domain via the bridge. At least some of the first domain is outputted. An apparatus includes a processor that defines first and second domains and a bridge that enables transfer of entities between domains, an entity identifier that identifies entities in the domains, a stimulus identifier that identifies stimuli, and a display that outputs at least some of the first domain. The processor transfers entities from first to second domain responsive to a first stimulus, and transfers entities from second to first domain responsive to a second stimulus.

The invention discloses a fault diagnosis method for industrial unbalanced fault data and related equipment. The method comprises the following steps: constructing a basic fault diagnosis model containing a double-branch feature extraction network, a sub-domain adaptive alignment module and a classifier; fusing actual and simulated vibration data with labels to generate a source domain training set, and collecting identically distributed label-free data as a target domain sample set; inputting a double-branch network to extract features, calculating feature distribution difference loss and main classification loss, and performing joint optimization training to obtain a target fault diagnosis model; and inputting a to-be-diagnosed vibration signal and outputting a fault diagnosis result. Therefore, the problem of insufficient single-scale feature representation is solved through double-branch feature fusion, the problem of cross-working-condition inter-domain offset is eliminated by using sub-domain distribution alignment, the problem of extreme imbalance of samples is relieved based on simulation data expansion, and the industrial fault diagnosis precision and the cross-domain reliability are further remarkably improved.