Patents
Literature
Patsnap Eureka AI that helps you search prior art, draft patents, and assess FTO risks, powered by patent and scientific literature data.

16results about How to "Effective migration" patented technology

Cross-modal brain tumor image segmentation method with finite source domain label

The invention relates to a cross-modal brain tumor image segmentation method with a finite source domain label, which is used for carrying out cross-modal image segmentation by effectively utilizing unmarked data of a source domain and a target domain under the condition that source domain labels are scarce aiming at a multi-modal brain tumor data set. The method comprises the following specific steps: constructing a source domain segmentation network, carrying out multi-scale consistency regular constraint, constructing a cross-modal adaptive network, carrying out double-circulation feature alignment through two sub-networks, then integrating the source domain segmentation network into the cross-modal adaptive network, and carrying out multi-scale consistency regular constraint on the cross-modal adaptive network. And forming a joint segmentation framework of a semi-supervised source domain and an unsupervised target domain, performing cross-modal adaptive segmentation model training optimization by using an end-to-end training mode, and testing a new target domain brain tumor image to obtain a segmentation result. The method can effectively reduce the domain difference, alleviates the problem of source domain label scarcity, and can be applied to the field of brain tumor cross-modal image segmentation.
Owner:NANJING DRUM TOWER HOSPITAL

Composite solid electrolyte diaphragm, preparation method and lithium ion battery

ActiveCN119542675AImprove liquid absorption capacityeffective migrationSecondary cellsCell component detailsSolid state electrolyteElectrical battery
The invention discloses a composite solid electrolyte diaphragm which comprises a bacterial cellulose membrane, fluorinated metal modified PEO electrolyte particles and functional group modified particles, and the fluorinated metal modified PEO electrolyte particles and the functional group modified particles are loaded in a three-dimensional network structure of the bacterial cellulose membrane. The functional group modified particles are functional group modified silicon dioxide particles and functional group modified LATP particles, the functional group modified silicon dioxide particles uniformly wrap the functional group modified LATP particles, and fluorinated metal of the fluorinated metal modified PEO electrolyte particles is combined with functional groups of the functional group modified particles through coordinate bonds. The invention also discloses a preparation method of the electrolyte diaphragm and a battery with the electrolyte diaphragm. The ionic conductivity of the electrolyte diaphragm is improved.
Owner:CHANGSHU INSTITUTE OF TECHNOLOGY

A cross-limb migration exoskeleton control method, device, equipment, medium and product

PendingCN122723687AEffective migrationefficient decoding
The application discloses a cross-limb migration exoskeleton control method, device, equipment, medium and product, relates to the exoskeleton control field, and the method comprises the following steps: acquiring donor limb training data and target limb data, inputting the donor input sample and the target input sample obtained after pretreatment into a cross-limb migration decoding model; the cross-limb migration decoding model carries out channel alignment on the donor input sample and the target input sample on a common lead set, constructs the latent space features of the two samples, and predicts the donor motion intention prediction result according to the donor latent space feature; the cross-limb migration decoding model is trained by using a total loss function; the real-time bioelectric signal data stream of the target limb is input into the trained cross-limb migration decoding model, the motion intention category, the corresponding confidence and the timing consistency index are obtained; an exoskeleton control instruction is output based on the confidence and the timing consistency index, and the motion intention prediction accuracy is improved.
Owner:BEIHANG UNIV

Composite photocatalyst as well as preparation method and application thereof

The invention provides a composite photocatalyst as well as a preparation method and application thereof, and belongs to the technical field of photocatalysis. The composite photocatalyst provided by the invention comprises CuBi2O4, CuCo2O4 and BiOF, wherein the CuCo2O4 and the BiOF grow on the CuBi2O4 in situ. CuCo2O4 and BiOF are introduced to the surface of CuBi2O4, and a dual-interface structure is formed among the three components, so that a multi-heavy-load ion shunting channel is constructed, and the catalytic activity and catalytic efficiency of the composite photocatalyst are improved; meanwhile, CuCo2O4 and BiOF which grow in situ have a closer interface with CuBi2O4, efficient separation of current carriers is promoted, and the composite photocatalyst has excellent photocatalytic activity and high catalytic efficiency. The data of the embodiment shows that the degradation rate of the tetracycline hydrochloride can reach 90.3% after the tetracycline hydrochloride is catalytically degraded by the composite photocatalyst provided by the invention within 15 minutes.
Owner:TAIZHOU UNIV (ZHEJIANG)

A Dual-Graph Driven Supervised Modeling Approach for Multi-Source Domain Transfer

ActiveCN118586434BSolve the negative transfer problemMigration is validInternal combustion piston enginesBiological modelsEngineeringGraph neural networks
The present invention discloses a dual-graph driven supervised modeling method for multi-source domain migration, which is mainly a dual-graph neural network model based on invariant relationship alignment, specifically including: extracting cross-domain invariant relationships and domain-specific relationships between variables through a public-private graph construction module, extracting cross-domain transferable relationships through a public graph information alignment module, extracting and filtering domain-specific information between variables through a private graph information filtering module, and collaboratively realizing the extraction of invariant relationships and the migration of multi-source domain knowledge to a target domain through a public graph information alignment module and a private graph information filtering module. The present invention can solve the negative transfer problem caused by the integration and distribution differences of multi-source domain information in traditional transfer learning, and realize effective knowledge migration from the source domain to the target domain; it can be applied to industrial soft measurement tasks of time series data such as speech and industrial sensor parameters in various transfer learning scenarios, and has good learning effect and generalization ability.
Owner:ZHEJIANG ZHENENG ELECTRIC POWER CO LTD XIAOSHAN POWER PLANT +2

Preparation method of Zn0. 3Cd0. 7S (at) CFR heterojunction photocatalyst

The invention discloses a preparation method of a Zn0. 3Cd0. 7S (at) CFR heterojunction photocatalyst, which comprises the following steps: firstly, preparing a Zn0. 3Cd0. 7S catalyst by a hydrothermal method, and dispersing the Zn0. 3Cd0. 7S catalyst in ultrapure water; then, catechol, ammonia water and formaldehyde are sequentially added into the dispersion liquid, CFR evenly grows on the surface of Zn0. 3Cd0. 7S through a hydrothermal reaction, a coating layer is formed, and a final product is obtained after washing and drying. The preparation method is simple, the process is easy to control, and the prepared photocatalyst shows excellent H2O2 yield and good cycle stability under the condition that a sacrificial agent is not added.
Owner:YANSHAN UNIV

A mechanical equipment fault diagnosis method, device, equipment and readable storage medium

The application discloses a mechanical equipment fault diagnosis method, device and equipment and a readable storage medium. Known fault information multi-source frequency domain signals of same or same type mechanical equipment under a first working condition are acquired, and the known fault information multi-source frequency domain signals under the first working condition are taken as source domain data sets. Unknown fault information multi-source frequency domain signals of the mechanical equipment to be diagnosed under a second working condition are acquired, and the unknown fault information multi-source frequency domain signals under the second working condition are taken as target domain data sets. A multi-source dense adaptive adversarial network model is trained by using the source domain data sets and the target domain data sets, and a mechanical equipment fault diagnosis model is obtained. The real-time acquired multi-source frequency domain signals of the mechanical equipment to be diagnosed under the second working condition are input into the mechanical equipment fault diagnosis model, and a fault diagnosis result is output. The application can better utilize the multi-source information of the mechanical equipment in data level fusion and feature level fusion.
Owner:XI AN JIAOTONG UNIV

BNVO / VC MXene / RGO photo-thermal piezoelectric catalytic material as well as preparation method and application thereof

The invention provides a BixNdyVO4 / V4C3MXene / RGO photo-thermal piezoelectric catalytic material as well as a preparation method and application thereof, the photo-thermal piezoelectric catalytic material is a compound of BixNdyVO4, V4C3MXene and RGO, and the ratio of x to y is 1: (0.5-0.75). The preparation method comprises the following steps: S1, mixing a multi-layer V4C3MXene dispersion liquid with a graphene oxide dispersion liquid, adding a bismuth source and a neodymium source, and stirring to obtain a reaction precursor liquid; and S2, carrying out a hydrothermal reaction on the reaction precursor solution, and washing and drying the obtained product to obtain the BNVO / VC MXene / RGO photo-thermal piezoelectric catalyst. According to the invention, the BixNdyVO4 / V4C3MXene / RGO material is utilized to improve the reduction stability of the material and the selectivity of methane generation through dark / photo-thermal carbon dioxide hydrogenation reduction.
Owner:SHAANXI UNIV OF SCI & TECH

Microbe-mof photosensitive active material and preparation method and application thereof

PendingCN122303212AImprove photocatalytic activityImprove light absorption capacityMicroorganismMicrobiology
This invention belongs to the field of photocatalysis technology, specifically relating to a microbial-MOF photosensitizing material, its preparation method, and its application. The microbial-MOF photosensitizing material comprises antibiotic-degrading bacteria and a metal-organic framework shell encapsulating the surface of the antibiotic-degrading bacteria. The metal-organic framework in the microbial-MOF photosensitizing material provided by this invention has a porous structure, enabling the encapsulated microorganisms to receive sufficient light, enhancing light absorption capacity, and facilitating rapid carrier migration. Ultimately, this gives the microbial-MOF photosensitizing material superior photoelectrocatalytic activity, lowers the photoelectrocatalytic reaction energy barrier, and increases the kinetics of antibiotic degradation.
Owner:KUNMING UNIV OF SCI & TECH

Electrolytic tank for producing hydrogen by electrolyzing pure water

The invention relates to the technical field of electrolytic baths, and discloses an electrolytic bath for producing hydrogen by electrolyzing pure water, the electrolytic bath comprises an electrolytic bath main body, a plurality of diaphragms are detachably connected to the inner side of the electrolytic bath main body, a driving shaft is rotatably connected to through holes in the bottoms of the diaphragms, and a plurality of second gears are fixedly connected to the exterior of the driving shaft; and a second rack plate is slidably connected to the lower side of the exterior of the diaphragm, a connecting plate is further slidably connected to the lower side of the exterior of the diaphragm, cleaning plates are fixedly connected to one side of the top of the connecting plate and one side of the top of the second rack plate, and guide rods are slidably connected to through holes in the tops of the cleaning plates. Through mutual cooperative use of a cleaning plate, a connecting plate, a second gear, a second rack plate and other structures, impurities on the outer surface of the diaphragm can be automatically cleaned, deposition of the impurities of the diaphragm can be effectively reduced, it is ensured that produced hydrogen has higher purity, and the high requirements of industry and chemical reactions for hydrogen quality are met.
Owner:HUANGHUAI UNIV

A small sample target detection method based on support feature diffusion generation

This invention relates to the fields of image processing and artificial intelligence, and particularly to a few-shot object detection method based on support feature diffusion generation. The method includes steps such as constructing a few-shot object detection base class meta-learning training set and grouping image data within the training set. This invention introduces a support feature diffusion generation module, applying a diffusion model to the generation and optimization of support features. During the training phase, the inverse process of adding noise to features is learned, enabling the model to recover robust support feature representations from noise. During the testing phase, through progressive reverse diffusion, a more stable and diverse set of support features can be generated from a small number of support samples, effectively mitigating the feature instability and overfitting problems caused by few samples.
Owner:SOUTHWEST JIAOTONG UNIV

A method for constructing generation recommendation supervision fine-tuning data based on frozen semantic embedding guidance

The application discloses a kind of based on frozen semantic embedding guide's generative recommendation supervision fine-tuning data construction method.The application avoids the distortion of semantic space and noise interference of collaborative training by freezing the semantic embedding obtained externally as a stable anchor, effectively protecting the content priori knowledge. At the same time, on the basis of maintaining semantic consistency, the collaborative neighborhood and sequence transfer preservation loss are introduced, so that the item representation can still completely retain the behavior patterns and interest evolution law in the original collaborative filtering space during the alignment process to the semantic space, obtaining high-quality fusion representation with semantic enhancement and collaborative authenticity. Based on the fusion representation, it is discretized into multi-level semantic identifiers through residual clustering, and instruction-response samples suitable for large language model supervision fine-tuning are constructed, realizing the effective migration of traditional recommendation knowledge to generative recommendation paradigm, enabling large language model to have sequence recommendation capability, and significantly improving the recommendation effect in cold start scenario.
Owner:BEIJING NORMAL UNIVERSITY

A low-code integrated development system based on multi-path intelligent recommendation

ActiveCN115756477BImprove quality assurancereduce dependenceCode moduleUser Privilege
The application discloses a low-code integrated development system based on multi-path intelligent recommendation, stores code modules through a code module warehouse subsystem, and records quality information and version information of the code modules and user permission information, thereby reducing dependence of the system on business scenarios and improving usability and flexibility of the system. Modification and update of the code modules are realized through a low-code development subsystem; a multi-path intelligent recommendation algorithm of the code module warehouse subsystem is used to search and recommend code modules for the low-code development subsystem, selective elimination of custom code modules can be realized, hierarchical clustering of the code modules is realized, search and recommendation of the code modules are realized according to a quality level coefficient, and higher-quality code modules can be provided for users. Process tracking and quality assurance documents are established through an automatic process management subsystem, thereby improving quality assurance of the custom code modules and reducing process management overhead caused by introduction of the low-code development platform.
Owner:BEIJING INST OF TECH

A multi-stage domain adaptive railway overhead line intelligent inspection cloud-edge collaborative real-time detection method

This invention discloses a multi-level domain adaptive intelligent inspection method for railway catenary with cloud-edge collaborative real-time detection, belonging to the field of rail transit technology. The method includes generating a dynamic weight adjustment factor based on a cloud dataset; extracting image features from the cloud dataset and generating distilled features based on the dynamic weight adjustment factor; performing multi-scale compression on the image features of the cloud dataset to obtain distilled compressed features; generating environmental features based on the dynamic weight adjustment factor and distilled features; concatenating the environmental features and distilled compressed features to obtain cloud-based large-scale model features; obtaining weather features based on weather data; performing conditional convolution on edge data and concatenating sparse features to generate sparse features; concatenating the sparse features, weather features, and cloud-based large-scale model features to obtain fused features; and performing data transformation to obtain prediction results. This invention solves the problems of low detection accuracy, poor adaptability, and difficulty in achieving efficient cloud-edge collaborative real-time detection in existing technologies.
Owner:NANJING ZHILIANSEN INFORMATION TECH CO LTD

A pre-training language model continuous learning method based on general sub-network pruning

The application provides a pre-training language model continuous learning method based on general sub-network pruning. The method sets zero to redundant neurons in the network through general sub-network pruning, thereby compressing original knowledge, and then pre-trains and updates the idle neurons set to zero on new professional field data, performs field collaborative continued pre-training, increases the field knowledge contained in the model under the condition that the size of the model remains unchanged, and the different field knowledge can work collaboratively, thereby achieving the effect of knowledge transfer.
Owner:HUAWEI TECH CO LTD +1

Preparation method and application of all-inorganic perovskite nanocrystalline composite cellulose material

The invention discloses a preparation method and application of an all-inorganic perovskite nanocrystalline composite cellulose material, and relates to the technical field of inorganic semiconductor materials, the preparation method comprises the following steps: mixing cesium carbonate, octadecene and oleic acid to obtain a cesium oleate solution; mixing lead bromide and octadecene to obtain a lead bromide precursor solution; injecting oleic acid and oleylamine into the lead bromide precursor solution, then heating, adding cellulose after lead bromide is completely dissolved, and stirring at a constant rate to obtain a lead bromide precursor / cellulose composite material; and injecting a cesium oleate solution into the lead bromide precursor / cellulose composite material, and centrifuging to complete the preparation. The invention also provides application of the material in humidity sensing. The full-inorganic perovskite nanocrystalline composite cellulose material prepared by the method has good reversibility and repeatability, and a humidity sensor based on the material has the advantages of simple preparation, high sensitivity, quick response, high stability in long-term work and the like.
Owner:ZHEJIANG OPTICAL STORAGE & CHARGE ENERGY TECH CO LTD