21results about How to "Eliminate negative effects" patented technology

The utility model discloses a device for synchronously controlling a withdrawal and straightening machine and a dummy bar of a continuous casting machine. The device comprises a potentiometer, a PLC (Programmable Logic Controller) control system, a signal isolation module, a withdrawal and straightening machine frequency converter, a dummy bar frequency converter and a digital display instrument. The potentiometer is connected to an analog quantity input port of the PLC control system through the one-input one-output optoelectronic isolator and is used for inputting an initial control signal; the output end of the PLC control system is connected with the withdrawal and straightening machine frequency converter and the dummy bar frequency converter to drive the two frequency converters to operate. A feedback signal of the withdrawal and straightening machine frequency converter is divided into two paths to be transmitted to the PLC control system and the withdrawal and straightening machine digital display instrument through the one-input two-output isolator, and a feedback signal of the dummy bar frequency converter is directly accessed to the dummy bar digital display instrument. The device eliminates electromagnetic interference through a photoelectric isolation module, integrates a water distribution model and a molten steel temperature data interface, adopts a standardized interface design for hardware, is adaptive to different continuous casting machine models, has the characteristics of strong anti-interference capability, stable signal transmission and visual real-time monitoring, and effectively improves the operation reliability of equipment.

The coil component 100 has a coil with a shape that improves the overall performance L×Isat / DCR of the coil component. It has a spiral conductor portion (first spiral conductor portion 11) with a central axis O along a first direction and having multiple turns, and a lead-out conductor portion (first lead-out portion 14) extending from the outer peripheral end of the spiral conductor portion towards a second direction (X1-X2 direction) intersecting the first direction (Z1-Z2 direction); and a core body portion 30 containing magnetic powder and adhesive material, covering the spiral conductor portion and the lead-out conductor portion, and exposing the end face of the lead-out conductor portion (first lead-out portion end face 14E). The two sides of the lead-out conductor portion along the second direction (inner side 14i and outer side 14p) have non-parallel portions when viewed along the first direction.

To achieve a power storage device capable of suppressing a negative effect of water that has infiltrated an inside of a case on an electrical device. In a power storage device of one embodiment of the present disclosure, the case includes a frame including a column unit and a beam member that links the column units and is assembled with a fastening member, a partition member that is fixed to the frame with the fastening member and divides an inside of the case into a space in which the electrical device is accommodated and another space, and a cap-shaped member that is fixed to the frame. The partition member includes waterproof portions at both end portions thereof on the left and right. In a state where the partition member is arranged on the upper side with respect to the electrical device, the waterproof portions are covered with the cap-shaped member.

This invention provides a training method, localization method, and system for an intelligent agent localization model based on dialogue and vision collaboration. The training method is applied to a federated learning system and includes: a server initializing a global model and distributing it to a client; the client receiving model parameters, training the model using local dialogue commands and visual data, generating multimodal representations, and then inputting them into two structurally identical but parameter-independent localization branches for parallel processing, outputting localization representations and prediction results, and executing navigation decisions based on the localization results; constructing a total loss function that includes a loss term constraining the consistency of the bi-branch representations and constraining the consistency of navigation state changes, and updating the local model; the client evaluating reliability based on representation consistency and navigation stability, uploading the model update and reliability to the server, and the server using this data to weighted aggregate and update the global model, iterating until convergence. This invention can suppress noisy dialogue interference and improve localization accuracy, navigation stability, and model generalization ability in federated learning scenarios.

This invention provides a method for embedding wires in continuous fiber-reinforced composite materials and vertically leading them out, comprising the following steps: laying a first insulating protective layer on a first composite material matrix; arranging multiple wires at predetermined intervals and in parallel on the first insulating protective layer; laying a second insulating protective layer above the wires; encapsulating to form a composite material preform; laying a second composite material matrix on the second insulating protective layer; forming a prefabricated cavity in the second composite material matrix through layer-by-layer stacking; filling the prefabricated cavity with a placeholder structure; encapsulating to form a complete preform; and curing the complete preform. After curing, the placeholder structure is removed from the second composite material matrix, and the wires can be vertically led out and moved from the prefabricated cavity. This invention can reduce the negative impact of embedded wires on the mechanical properties of composite materials and effectively prevent leakage problems caused by damage to the insulating varnish layer on the wire surface.

This invention discloses a robust fusion method for large models based on semantically aligned fuzzy clustering. The method includes: first, obtaining a sequence of probability distribution vectors from the outputs of multiple heterogeneous large-scale pre-trained models for the samples to be processed; then, introducing non-negative reliability weights to weight this sequence to obtain an aggregated probability matrix, and constructing a graph Laplacian matrix accordingly; second, approximating the aggregated probability matrix into a fuzzy membership matrix and a semantic prototype matrix, constructing a cost objective function by combining the graph Laplacian matrix, and iteratively updating each matrix and weight until convergence; finally, solving for the maximum value of the converged fuzzy membership matrix to obtain the sample prediction category result. This invention can effectively suppress the influence of inferior models in unsupervised environments, significantly improving the accuracy and robustness of fusion prediction.

This invention discloses a polyphosphazene / MXene self-supporting electrode loaded with cobalt single atoms, its preparation method, and its application, belonging to the field of seawater uranium extraction technology. Ti3AlC2 powder is etched in an HCl-LiF mixed solution and then ultrasonically exfoliated to obtain Ti3C2T. x Nanosheet suspension; simultaneously, hexachlorocyclotriphosphazene, 4,4'-dihydroxydiphenyl sulfone, and cobalt salt were dissolved in a solvent, and triethylamine was added. A condensation reaction was then carried out under ultrasonication to obtain Co-PZS; then Ti3C2T x After the nanosheet suspension is uniformly mixed with Co-PZS and carbon fiber fabric, it is transferred to a hydrothermal reactor for hydrothermal reaction. The resulting hydrothermal reaction product is then dried and calcined to obtain a polyphosphazene / MXene self-supporting electrode loaded with cobalt single atoms. This electrode is used for the electrochemical adsorption of uranyl ions in the solution system and exhibits excellent comprehensive uranium extraction performance, with advantages such as high capacity, fast rate, strong selectivity, and good stability, meeting the application requirements of seawater uranium extraction.

The invention discloses a complex dynamic industrial visual positioning method and system based on multi-agent collaboration, and relates to the technical field of machine vision, distributed visual agent nodes are deployed in an industrial dynamic operation scene, and node microsecond-level time synchronization, space joint calibration and communication networking are completed; the nodes synchronously collect industrial images according to a preset time sequence, and a standardized data set is generated through preprocessing; dynamic target feature extraction is completed based on an improved deep convolutional neural network, and cross-node feature matching is completed in combination with epipolar constraints; a multi-view geometric constraint model is constructed to solve the three-dimensional pose of the target, and a positioning result is generated through dynamic interference compensation optimization; and outputting a positioning result, completing closed-loop feedback, and dynamically adjusting node parameters to realize multi-agent collaborative global optimization. According to the invention, the precision and the anti-interference capability of visual positioning in a complex dynamic industrial scene are improved, the multi-node load balancing is optimized, and the visual positioning operation requirements of various industrial dynamic scenes can be met.

The utility model discloses a kind of via hole structures of FPC preventing big impedance, belong to FPC board via hole field, a kind of via hole structure of FPC preventing big impedance, including FPC board, A line upper end, A line lower end, K line upper end, K line lower end are equipped on the FPC board, the A line upper end, K line upper end terminal are equipped with double via hole, the A line upper end is electrically connected with A line lower end by double via hole, the K line upper end is electrically connected with K line lower end by double via hole, wherein double via hole is specifically two via hole compositions, via hole outer diameter is specifically 0.5mm, via hole inner diameter is specifically 0.25mm, the hole spacing between two via holes is specifically 0.3mm;The utility model has the advantages of prior art in that: the number of via hole in the structure of FPC is increased to 2, avoid the risk of abnormal function caused by big impedance, reduce the burnout generated in the use process of product due to the via hole process problem of supplier, eliminate the negative influence caused by after-sales function bad, greatly reduce after-sales cost.

This invention relates to the field of tonic wine stabilization technology and discloses a nano-encapsulation method for stabilizing tonic wine. The method includes constructing a digital encapsulation matrix model of a tonic wine system containing multiple active components, extracting the molecular structural characteristics and physicochemical parameters of the active substances, and, in conjunction with preset encapsulation target requirements, analyzing the key limiting factors for the stability of the tonic wine system and generating an encapsulation demand vector. Then, based on this vector, a multi-level screening operation is performed in the model to locate the target encapsulation domain. Within this domain, relevant data of the nanocarrier material are integrated, and the compatibility between it and the active substance units is evaluated using multi-parameter fusion calculation rules to obtain an optimal sequence of nano-encapsulation materials. Based on this sequence, a nano-encapsulation operation instruction set is generated to perform nano-encapsulation processing on the tonic wine system, obtaining an encapsulated composite system. The dispersion uniformity and sedimentation rate of the system are monitored in real time, and its stability is verified using preset standards, outputting verification result data.

This invention belongs to the field of microbial preparation technology, specifically relating to a biocontrol agent composition and its application in improving growth barriers caused by high-temperature soil sterilization and controlling plant diseases. The technical problem this invention aims to solve is to provide a new option for alleviating soil obstacles caused by high-temperature sterilization. The technical solution of this invention is a *Pseudomonas aeruginosa* 115-17, with the preservation number CCTCC No: M2026293. This invention provides a novel *Pseudomonas aeruginosa* 115-17 strain, which possesses nitrogen-fixing, inorganic phosphorus-solubilizing, and siderophore-producing abilities; it can also promote pepper growth, control plant diseases (mainly pepper blight), and improve plant growth inhibition caused by high-temperature soil sterilization. This invention also provides a combination of Bacillus velezensis Bv-6 (CCTCC No: M20191106) and newly isolated Pseudomonas aeruginosa 115-17 to form a microbial inoculum, thereby effectively alleviating plant growth obstacles caused by high-temperature sterilization and effectively preventing and controlling pepper blight.

The application provides a cross-domain fault diagnosis method based on uncertainty quantification and differential collaborative optimization, and relates to the technical field of intelligent fault diagnosis. A source domain neural network model with uncertainty quantification capability is obtained by training a target neural network model; the parameters of the classification module of the source domain neural network model are frozen, the parameters of the feature encoding module of the source domain neural network model are updated, the samples in the target domain data set are double-checked, and are divided into a reliable sample set and an unreliable sample set; based on the sample division result, the source domain neural network model is differentially collaboratively optimized to obtain a trained target network model, and the target network model is used for fault diagnosis. The application can effectively suppress the negative influence of noise pseudo labels, realize fine alignment of features on the premise of protecting the privacy of source domain data, and improve the diagnosis precision of mechanical equipment faults under varying working conditions.

This invention specifically relates to a distributed power management strategy for a hybrid electric propulsion system for electric aircraft, belonging to the field of aviation electrification. The system consists of a fuel cell, an energy storage battery, and a supercapacitor connected in parallel to a DC bus via a power converter. In this invention, each unit is independently controlled based on local information, eliminating the need for a communication network: the fuel cell adaptively switches between voltage and power modes according to the load power and battery state of charge, avoiding heavy or light load operation; the battery adjusts its steady-state operating point through a virtual resistance droop mechanism, absorbing steady-state power deficits or excess power; the supercapacitor absorbs dynamic power through a virtual capacitance droop mechanism, suppressing load fluctuations. This invention effectively suppresses the negative impacts of heavy / light loads, load fluctuations, and frequent start-stop cycles on fuel cell lifespan and efficiency, reduces hydrogen consumption, and improves overall system reliability.

This invention discloses a robust denoising method for wind turbine data based on information theory learning. The method includes: acquiring historical wind turbine data under different operating conditions; constructing an autoencoder model containing a cascaded encoder and decoder; constructing a denoising loss function and a feature regularization loss function; directly using the acquired historical wind turbine data as training data, weighting the denoising loss function and the feature regularization loss function to construct the total loss function of the autoencoder model, and iteratively training the autoencoder model using the training data itself as the reconstruction target; inputting the wind turbine data to be denoised into the trained autoencoder model for processing to obtain the denoised wind turbine data. This invention's method does not require the use of clean, noise-free data during training, and the model training process exhibits good noise robustness, enabling effective denoising of wind turbine data.

The utility model relates to hot melting device field discloses a pearl cotton waste material collection and recycling device, including the crusher, hot melting subassembly and waste gas treatment box, the crusher bottom is equipped with the discharge gate, is equipped with the feeding hopper on the crusher, the hot melting frame inner chamber is equipped with the material guiding cylinder, is equipped with the material receiving box on the material guiding cylinder, the discharge gate is connected into the material receiving box, the material guiding cylinder output is equipped with the hot melting cylinder, is equipped with a plurality of gas outlets on the hot melting cylinder, is equipped with the gas guide hose on the gas outlet, the gas guide hose all are connected into the waste gas treatment box. The utility model can cut and crush the pearl cotton to complete to handle by oneself, and utilize the relative installation helical frame to make it melt to press into the hot melting cylinder continuously, then add target additive to the material, get target product and extrude, organic waste gas flows into the waste gas treatment box and photocatalytic decomposition, remove the negative influence that waste gas emission causes to the environment, improve production use experience.

The invention discloses a long-term semantic VSLAM method and system for dynamic accumulated noise suppression and ghosting elimination, and relates to the technical field of robot visual navigation and map construction, and the method comprises the following steps: collecting a current frame image in the continuous inspection process of a robot, and carrying out the processing of the current frame in the processing of the current frame; semantic mask extraction based on the instance segmentation model and dense optical flow field calculation based on the current frame and the previous key frame are executed in parallel; according to the method, potential dynamic object categories and accurate contours in the images can be recognized by adopting the YOLACT instance segmentation model, the large displacement situation possibly brought by the motion of the robot is coped with by combining a parallel execution optical flow estimation algorithm, semantic understanding of the environment and accurate synchronous capture of motion information are achieved, and the method has the advantages of being high in practicability and high in practicability. And the optical flow generated by the motion of the object and the background optical flow caused by the motion of the camera are effectively distinguished by measuring and analyzing the direction consistency of the optical flow vector, so that the accuracy and robustness of dynamic judgment are improved.

The application belongs to the field of materials, and particularly relates to a crystal structure generation method based on a diffusion model and a self-attention attenuation graph network, aiming to solve the problems of low matching degree of crystal structure generation and target properties and limited prediction accuracy. The method comprises the following steps: obtaining real crystal structure data, and performing feature extraction to construct a crystal structure dataset; inputting the crystal structure dataset into a diffusion model for training under the condition of target material properties; converting the crystal structure dataset into a crystal graph, and inputting the crystal graph into a graph neural network for training; in the message passing process, combining a multi-head self-attention mechanism and a weight attenuation function based on atomic distance to update node features and edge features, so as to obtain a prediction model; inputting target material properties into the trained diffusion model to generate candidate crystal structures; inputting the candidate crystal structures into the trained prediction model to output property prediction values; comparing the property prediction values with the target material properties, and screening to obtain a target crystal structure.

The application provides a conductive grease composition and a preparation method thereof. The conductive grease composition comprises the following components in parts by mass: base oil 80-85 parts, thickening agent 15-20 parts, antioxidant 2-4 parts, antirust agent 1-3 parts, and conductive agent 2-3 parts; the conductive agent is an ionic liquid and carbon black in a mass ratio of 1:1-2. The conductive grease composition provided by the application has good lubricating performance and excellent and stable conductivity, can obviously reduce the resistivity of the grease, and effectively alleviates the bearing electric corrosion problem.

A cognitive impairment auxiliary analysis method based on multi-view brain network feature fusion, belonging to the field of medical information processing technology, solves the technical problem that existing technologies struggle to fully exploit the correlations and complementarities between multi-level structural features of patients' brain functional connectivity data, leading to low accuracy in assessing patients' brain functional status. This method uses rs-fMRI as input data, extracts multi-level brain connectivity information by constructing functional connectivity networks and higher-order functional connectivity networks, and builds a multi-view local-global graph neural network model (MVLG-GNN) to achieve collaborative modeling, adaptive fusion, and auxiliary analysis combining group relationships of subjects from multiple views of brain connectivity features. This invention is also applicable to the field of Alzheimer's disease brain data processing.

An optical device may include, for example, a base (21a); ​​a light-emitting element (32) disposed on the base (21a) and emitting laser light; a plurality of optical components disposed on the base (21a) and guiding the laser light emitted from the light-emitting element (32) to an optical fiber (107) and coupling it to the optical fiber (107); and a shielding portion (101d1) disposed on the base (21a) and shielding stray light that deviates from a predetermined optical path and is directed toward a second optical component, which is also an optical component, within a first optical component, and reflecting the stray light in a direction deviating from the first optical component. In the optical device, the shielding portion (101d1) may also reflect stray light away from the base (21a).

The application provides a preset time disturbance observation production capacity control method and device of a reentrant manufacturing system, and relates to the technical field of control science and engineering. In the product manufacturing process of driving the reentrant manufacturing system according to the specified product delivery time limit and the specified system delivery output, the system disturbance observer is used to accurately estimate the external disturbance of the system within the preset disturbance estimation time period before the specified product delivery time limit, and the state feedback control mechanism is used to simultaneously consider the system output control function and the external disturbance influence elimination function, so that the corresponding reentrant manufacturing system can respond to market demand within the specified product delivery time limit, complete the product delivery work of the specified system delivery output, effectively eliminate the negative influence of the external disturbance of the system on the system capacity, and improve the system reliability and control robustness of the reentrant manufacturing system.