17results about How to "Verify superiority" patented technology

This invention relates to the field of multimodal sentiment analysis technology, and in particular to a method and system for multimodal sentiment analysis based on sequential translation under uncertain and missing modalities. The method includes: acquiring multimodal sentiment data, including visual, textual, and auditory modalities; constructing a multimodal sentiment analysis model based on sequential translation, firstly by extracting self-attention encoding features of each modality using a Transformer encoder; and then designing a bidirectional sequential translation architecture centered on the textual modality. This invention improves the utilization of textual modal features in the feature fusion process, thereby enhancing the quality of multimodal fusion features and providing strong support for subsequent complex sentiment analysis.

This invention discloses a self-supervised heterogeneous graph learning method for item recommendation based on a unified spatial-spectral perspective. This method is a self-supervised heterogeneous graph learning framework that simultaneously considers spatial and spectral domain information, which is beneficial for graph representation learning. Specifically, a spatial encoder is used to aggregate information from local neighbor nodes, and a spectral encoder is used to extract global structural information, thereby suppressing noise from local aggregation. Subsequently, a homeomorphism mapping method is proposed, which can align dual-domain representations while avoiding the loss of domain-specific information. Furthermore, a spectral optimization module is proposed, which uses fused representations to generate cluster-level guidance for effectively optimizing the original representations. Extensive experimental results on various real-world heterogeneous graph datasets for item recommendation demonstrate the significant superiority of this invention.

This invention provides a fluorinated DSPE-PEG material, its preparation method, and its applications, belonging to the fields of nanomedicine delivery systems and biomedicine. This invention prepares fluorinated DSPE-PEG materials with different fluorination modification rates by fluorinating distearate phosphatidylethanolamine polyethylene glycol amino groups, and uses these materials as carriers to load sangxin (a type of glycerol) to form fluorinated nanomicelles. This invention systematically screens the effect of fluorination modification rate on the performance of the nanomicelles, determines the optimal fluorination modification rate, and confirms that it can overcome the defect of low cellular uptake of PEGylated carriers and enhance the anti-colorectal cancer activity of the loaded drugs. The fluorinated nanomicelles have uniform particle size, good stability, and excellent biocompatibility with the carrier material. This invention provides a new strategy for the efficient delivery of small molecule active ingredients from traditional Chinese medicine and the treatment of colorectal cancer.

The utility model discloses a kind of concave composite chamfering cutter, belong to machining tool design technical field, the composite chamfering cutter includes two center symmetrical cutter teeth, cutter tooth is composed of back rake face, oblique front and cylindrical arc surface, wherein: the included angle range of back rake face and vertical face is 35 ° to 50 °, the included angle range of oblique front and vertical face is 8 ° to 15 °, cylindrical arc surface is the extension of cylindrical cutter body, the top end of cutter tooth is also provided with tooth end face, the included angle range of tooth end face and horizontal plane is 3 ° to 8 °. Compared with prior art, the composite chamfering cutter carries out mirror image processing to the two blades of tool, and designs specific cutter tooth angle, and the oblique front of specific angle can effectively increase the strength of tool, facilitate chip removal, and the back rake face of specific degree can reduce the wear of cutting edge while reducing vibration;The stability and service life of tool are improved, and the overall tool can effectively reduce the vibration interference in the machining process during the machining process, to ensure the stability and consistency of machining quality.

The present application relates to a kind of atomizers, including the inertial sensor and signal processor arranged on atomizer, the inertial sensor, signal processor are connected with the drive circuit of atomizer;The signal processor is based on the attitude signal collected by the inertial sensor, it is handled into the regulation and control signal that atomizer can directly respond, the drive circuit real-time adjustment the output of atomizer.The present application also relates to the calibration method of the atomizer.The atomizer of the present application significantly improves the inhalation deposition efficiency and inhalable fraction of atomized drug.

This invention proposes a generalizable federated learning method and system based on perceptual heterogeneity features and consistency of the fundamental matrix. The system includes a central server and multiple clients. The central server has an unlabeled public image dataset, and each client holds a labeled private image dataset and deploys a local image classification model that can be decomposed into an image feature extractor and a classifier. Each client is equipped with a feature mapping layer that projects the image features output by the image feature extractor onto a unified high-dimensional feature space. This invention solves the heterogeneity and catastrophic forgetting problems through the FBCL federated learning framework.

The invention discloses a modulation signal identification method based on convolution attention and multi-dimensional feature fusion, and solves the problems that in the prior art, different signals in the same signal family always show similar features under the condition of a low signal-to-noise ratio, so that a model is difficult to distinguish effectively, the identification accuracy is reduced, and the identification accuracy is low. The modulation signal identification under a low signal-to-noise ratio is realized, the identification precision is higher, and the calculation complexity is lower; the method comprises the following steps: acquiring an original complex signal, normalizing the original complex signal to obtain an IQ sequence, and obtaining an AP sequence through the IQ sequence; then, a pre-trained signal identification network is used for identification; a preprocessing module converts the input IQ and AP sequences into IQ and AP feature maps; the feature extraction module extracts local and global features to form a path feature map; the WAFF module performs weighted fusion on the two paths of features; and finally, the classification head outputs a classification result according to the fusion features.

The application relates to a water surface unmanned ship improved GA-SA path planning method for offshore wind farm inspection, and relates to a water surface unmanned ship path planning method.The application aims to solve the problems of low efficiency and low precision of an existing algorithm.The process is as follows: one, initializing algorithm parameters;two, performing chromosome coding of feasible solutions, and setting a fitness value function;three, taking the highest fitness value as the optimal individual;four, judging whether the convergence condition is met, if yes, outputting the current optimal individual, otherwise, entering five;five, taking the current optimal value as an initial solution of the SA algorithm, and randomly obtaining an updated solution;judging the energy change of the updated solution;if the energy change is less than 0, the updated solution is taken as the latest solution;otherwise, the updated solution is reserved according to a probability;six, judging whether the convergence condition is met, if no, repeating five;if yes, returning to three to replace the optimal individual, and repeatedly executing four to six until the convergence condition is met.The application is used in the field of path planning of water surface unmanned ships.

The present application proposes a stress inversion and numerical optimization based intrinsic strain field calibration method and system, belonging to the technical field of metal additive manufacturing numerical simulation. The present application solves the problems in the existing strain field calibration method, such as the lack of feedback calibration mechanism based on measured stress, the focus on residual deformation only, the inability to verify the material constitutive model, the poor generalization of intrinsic strain parameters, and the significant deviation between the prediction results of complex components and the actual measurement. The method comprises the following steps: S1: preparing a calibration sample according to the target material; S2: obtaining the measured value of residual stress at the measurement point of the calibration sample; S3: establishing a finite element model of the calibration sample; S4: establishing a target function and performing automatic iterative solution to obtain the optimal solution of the design variable. The system comprises a calibration sample preparation module, a stress measurement module, a modeling module and a design variable solving module.

The present application provides a kind of trace thiabendazole detection method and system of terahertz metamaterial resonance enhancement, the method comprises: based on the characteristic peak frequency of single period structure and thiabendazole, the parameter information of the metamaterial sensor is constructed, and modeling simulation is carried out according to parameter information, to obtain the terahertz metamaterial with L type composite double-peak structure;Preparation thiabendazole sample, and utilize terahertz metamaterial to carry out spectrum collection to thiabendazole sample, to obtain corresponding spectral data;Support vector regression model is constructed, and the optimization algorithm of weighted vector average is integrated, and global optimization of hyperparameter is carried out, to construct corresponding optimization model;According to spectral data and optimization model, corresponding detection model is constructed, and nondestructive testing of trace thiabendazole is realized using detection model.The present application not only realizes high sensitive quantitative analysis to trace thiabendazole residue, also verifies the feasibility and superiority of INFO optimization algorithm in the field of spectral detection.

The present application relates to a kind of hydraulic swing joint position servo system and its NDOB-SMC control method, first design hydraulic swing joint position servo system, servo valve and the cylinder body of hydraulic cylinder are integrated direct connection design in this system, abandon oil pipe connection, not only structure is more compact, and reduce part interference factor;Second, design a kind of sliding mode control strategy based on nonlinear disturbance observer, combine nonlinear disturbance observer and sliding mode control, give full play to respective advantages, improve the tracking accuracy and response speed of hydraulic swing joint;Finally, the component model library of AMEsim platform is rich and the powerful numerical operation ability of Matlab platform, by creating S-Function interface, the joint simulation of two platforms is realized, and the effectiveness and superiority of the proposed control method are verified.

The invention discloses an improved opportunistic network coding model for dynamically selecting coding, which is used for carrying out real-time channel evaluation and cache management improvement on opportunistic network coding, and selecting an optimal coding strategy according to evaluated channel state information and a cache state, so that the transmission performance in a complex multi-user and multi-relay environment is improved. Wherein channel quality during forwarding of different data packets is evaluated through channel gains, and corresponding appropriate data packets with high channel quality are selected to be transmitted to the next process; in the improvement of cache management, the data packet with the most potential stored in the cache is selected to be coded as the optimal coding strategy, and the coding strategy of the relay node is dynamically adjusted through real-time channel evaluation and cache management, so that the transmission efficiency in a complex wireless network environment is improved, the bit error rate is reduced, and the resource utilization rate is optimized.

The invention belongs to the field of intelligent biological prediction, and particularly relates to an RNA-small molecule binding affinity prediction method based on multi-view learning. The method comprises three stages of data set up and down sampling processing, feature construction and extraction based on RNA and small molecule sequences, and binding affinity prediction. According to the method, data set distribution is optimized through an up-down sampling method, multi-view feature information is constructed through sequence information of RNA and small molecules, a deep network is established to extract and depth biological information features, and finally, an MLP regression device is adopted to predict an RNA-small molecule binding affinity value, so that drug discovery work is assisted to be achieved. In order to solve the problem of imbalance of the existing data set, the invention provides a method of combining up and down sampling for processing, and the problem of imbalance is effectively relieved.

The application discloses a kind of image semantic transmission and multi-outlet DNN inference collaborative design method for star-ground fusion network, constructs an image semantic transmission and multi-outlet DNN inference collaborative processing model under the scene of star-ground fusion network, model includes the LEO satellite equipped with semantic encoder, at least one ground station equipped with semantic decoder and multi-outlet DNN inference module.Satellite is collected after remote sensing image data, by encoder, the semantic features extracted are transmitted to ground station, ground station reconstructs image by decoder, and executes multi-outlet DNN inference task, adopts early return mechanism in inference process, and when meeting inference accuracy threshold, inference result is output.On the basis of considering semantic recovery quality and inference accuracy, jointly satellite launch power, semantic compression rate, EE decision and computing resource allocation, total time delay of task completion is minimized, and image semantic transmission and multi-outlet DNN inference collaborative design are realized.

The application provides a kind of mobile robot navigation method based on deep reinforcement learning, realizes guiding to accelerate convergence before the training of deep reinforcement learning strategy, so that mobile robot can complete safe and efficient autonomous navigation only relying on on-board sensor in complex dynamic environment.The application comprises the following steps: S1, collecting data online, comprising the following steps: S11, training navigation model in simulation environment;Define the state space of mobile robot in simulation environment, and form state vector;S12, send state vector into graph attention network, obtain state representation;S13, policy network takes state representation as input, and outputs control action;The reference action of linear velocity and angular velocity is calculated by artificial potential field method;S2, during the training phase of navigation model, train critic network, policy network and graph attention network;S3, after the training of navigation model is completed, deploy navigation model on mobile robot.

The invention discloses a method for preparing 2-methylfuran by taking biomass-derived furfural as a raw material, and belongs to the technical field of furfural hydrodeoxygenation catalysis. Comprising a carrier B and an active component A loaded on the carrier B, the active component A comprises Cu and a co-catalytic metal; the co-catalytic metal comprises at least one of Ni, Co, Fe, Mo, W, Re and Nb; the carrier B is an Al-doped molecular sieve, and the molecular sieve is one of MCM-41, SBA-15, HZSM-5 and ZSM-5. And by taking a mixed solution of furfural and an alcohol hydrogen donor as a raw material, carrying out gas-phase hydrodeoxygenation reaction under the catalytic action of the bifunctional catalyst to prepare the 2-methylfuran. The process has the characteristics of low reaction temperature, low energy consumption, high safety, good selectivity, excellent catalyst stability and the like, and is suitable for green preparation of sustainable aviation fuel and high-energy-density biofuel.

This invention relates to the field of solar cell technology, specifically to a hole transport layer containing betaine hydrochloride additive and its inverse wide-bandgap perovskite solar cell. The hole transport layer comprises a nickel oxide layer and a self-assembled molecular layer formed on its surface. The self-assembled molecular layer contains (4-(3,6-dimethyl-9H-carbazole-9-yl)butyl)phosphonic acid and betaine hydrochloride. The introduction of betaine hydrochloride improves the conductivity of the hole transport layer through the interaction of its carboxylic acid groups with nickel oxide, and inhibits its clustering behavior through interaction with the self-assembled molecules, thereby improving film uniformity and effectively passivating buried interface defects in the perovskite light-absorbing layer. The inverse wide-bandgap perovskite solar cell fabricated based on this hole transport layer exhibits significantly improved open-circuit voltage and fill factor, ultimately achieving a photoelectric conversion efficiency exceeding 21.7%. This invention provides an effective interface engineering strategy for fabricating efficient and stable wide-bandgap perovskite solar cells and tandem devices.