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

51results about How to "Obvious performance advantage" patented technology

Autonomous robot decision-making system based on multi-modal perception fusion and method thereof

The invention relates to the technical field of autonomous robot decision-making systems, in particular to an autonomous robot decision-making system and method based on multi-modal sensing fusion, and the system comprises a multi-modal data collection module which is used for obtaining multi-modal sensing data of an environment; the data preprocessing module is connected with the multi-modal data acquisition module and is used for preprocessing the multi-modal sensing data; the feature extraction module is connected with the data preprocessing module and used for extracting features from the preprocessed multi-modal sensing data; the dynamic space-time diagram construction module is connected with the feature extraction module and is used for constructing a dynamic space-time diagram based on the features; the multi-modal data fusion module is connected with the dynamic space-time diagram construction module and is used for fusing the multi-modal data in the dynamic space-time diagram; through adaptive time alignment and a cross-modal space attention mechanism, the system can effectively process space-time inconsistency of data of different sensors, and centimeter-level positioning precision is realized.
Owner:LONGYAN LITAI INFORMATION TECH CO LTD

6G conduction signal fusion method, device and system for space-air-ground integration

The invention relates to a 6G conduction signal fusion method, device and system for space-air-ground integration. The method comprises the following steps: generating a conduction fusion signal at a transmitting end; modulating the conduction fusion signal from a time delay-Doppler domain to a time delay-time domain, and performing parallel-serial conversion to obtain a time domain conduction fusion signal; quickly capturing a satellite at a receiving end; after a satellite is captured, a received time domain conduction fusion signal is converted and restored to obtain an accurately estimated fractional Doppler value and fractional time delay, a channel matrix is reconstructed, data demodulation is performed by using the channel matrix, and navigation message information of the captured satellite is obtained; doppler positioning is carried out by using the fractional Doppler value and satellite information to obtain coarse positioning estimation of a receiver; and according to the four groups of navigation message information and the four groups of pseudo-ranges, Doppler positioning correction values are calculated to correct coarse positioning estimation of the receiver, and 6G conduction signal fusion transmission is realized. The method can improve the signal spectrum efficiency and the satellite utilization rate.
Owner:NAT UNIV OF DEFENSE TECH

Offline writer identification method and device, electronic equipment and medium

The invention discloses an offline writer identification method and device, electronic equipment and a medium. The method comprises the following steps: acquiring handwritten word image data; converting the handwritten image into a grayscale image, performing binarization processing on the grayscale image, adjusting the size of the processed image to a preset size, and constructing a training set; constructing a handwriting recognition model, and training the handwriting recognition model by adopting the training set; wherein the handwriting recognition model comprises a local modeling module, an identical branch and a global module, the local modeling module is used for capturing local features, the identical branch is used for extracting global features, and the global module is used for integrating the local features and the global features; and inputting a handwritten image to be recognized into the trained handwriting recognition model to obtain a result of a final classification layer of the model, performing softmax probability calculation, and taking a writer with the maximum probability as a recognition result. According to the method, robust feature representation can be effectively extracted from limited handwriting, and the recognition precision far higher than that of other methods is achieved.
Owner:SOUTH CHINA UNIV OF TECH

Small sample ancient character recognition method based on structural features and evolution modes

The invention discloses a small sample ancient character recognition method based on structural features and an evolution mode, and the method comprises the steps: S1, collecting an ancient character image, constructing a data set, and carrying out the data enhancement processing of the data set; s2, training a neural network by using the data set after data enhancement to obtain a small sample ancient character recognition model; and S3, inputting the target ancient character image into the small sample ancient character recognition model, and outputting an ancient character recognition result. According to the method, global features, local features and structural features and evolution rules of Chinese characters are considered in the process of identifying the ancient characters, and feature extraction of the ancient characters is innovative; according to the method, softer data enhancement is carried out on a data set with a small data size, the influence caused by data scarcity can be relieved, meanwhile, the generalization ability of the whole model is improved, and the problem of ancient text data scarcity is well solved.
Owner:SICHUAN UNIV

Large-scale MIMO-OTFS beam forming design method based on deep learning

The invention relates to a large-scale MIMO-OTFS beam forming design method based on deep learning, and belongs to the technical field of wireless communication. An optimized beamforming neural network (OBNN) is designed, a beamforming matrix of an MIMO-OTFS system is designed by taking maximization of spectrum efficiency as an optimization target, a base station modulates a time delay Doppler domain signal through OTFS and then sends the time delay Doppler domain signal through a large-scale antenna, a user carries out channel estimation based on a pilot symbol at an OTFS receiving end, a channel estimation value and an SNR value are taken as inputs of the OBNN, and an MIMO-OTFS system beamforming matrix is obtained. Model learning is carried out through minimizing a loss function, so that an optimal beam former weight is obtained. The method has the advantages that high robustness of OTFS modulation to Doppler frequency shift is utilized, the method adapts to a high-mobility communication environment, it can be ensured that a loss function is converged to local optimum through deep learning, and compared with other neural network methods, the spectrum efficiency of the method is remarkably improved by utilizing pre-training weight to adapt to channel estimation errors.
Owner:JILIN UNIVERSITY

Gas-sensitive sensor for detecting dissolved gas in oil, array of gas-sensitive sensor and preparation method of gas-sensitive sensor

The invention discloses a gas sensor for detecting dissolved gas in oil, an array of the gas sensor and a preparation method of the gas sensor. The gas sensor comprises a substrate, an interdigital electrode deposited on the surface of the substrate, and a p-n junction modified TMDCs gas sensitive material attached to the surface of the interdigital electrode, the p-n junction modified TMDCs gas sensitive material is a MoS2-WS2 heterostructure material, and the molar ratio of W to Mo in the MoS2-WS2 heterostructure material is (1.3: 1) to (1.7: 1). The MoS2-WS2 heterostructure is formed by adjusting the ion proportion of W and Mo, and the sensitivity and selectivity of the sensor to gas dissolved in oil are remarkably improved.
Owner:ELECTRIC POWER RESEARCH INSTITUTE OF STATE GRID NINGXIA ELECTRIC POWER COMPANY +2

Halton sequence optimization method and system based on RISC-V vector extension and application

The invention relates to a Halton sequence optimization method and system based on RISC-V vector extension and application. The Halton sequence optimization method comprises the following steps: 1) identifying mathematical characteristics generated by a Halton sequence; a Halton sequence is generated; after continuous index values are partitioned and grouped, simultaneous processing of the multiple index values is achieved through a parallelized grouping calculation method in combination with vectorization operation; 2) carrying out vectorization optimization by using an RISC-V vector instruction; in a Halton sequence generation process, carrying out parallel processing on the grouped index values by using an RISC-V vector extension instruction; 3) packaging into a dynamic link library; and 4) integration and calling with a Hadoop framework. According to the Halton sequence generation method, the number of loop iterations is remarkably reduced, and the Halton sequence generation efficiency is greatly improved. According to the method, resource waste in traditional serial computing is avoided, meanwhile, branch jump and condition judgment expenses are reduced, and the utilization rate of hardware computing resources is increased.
Owner:SHANDONG UNIV

Polyurethane wear-resistant and corrosion-resistant film for front edge of wind power blade and preparation method thereof

The invention relates to a wear-resistant and corrosion-resistant polyurethane film for the front edge of a wind power blade and a preparation method thereof, and belongs to the technical field of films. The dispersion and binding force of graphite-phase carbon nitride in a waterborne polyurethane emulsion are remarkably improved through ultrasonic-assisted solvent stripping, the dispersion and interfacial compatibility are further optimized by introducing carboxymethyl chitosan, and a composite film is formed on the surface of a matrix, so that corrosive ions are effectively shielded, the corrosion resistance is enhanced, and the service life is prolonged. The photocatalytic performance of the graphite-phase carbon nitride can also endow the film with self-cleaning and anti-photoaging capabilities. Besides, the carbon nanotubes and the silicon dioxide are compounded by adopting an in-situ hydrolysis technology, so that the interaction force between the carbon nanotubes and the silicon dioxide is remarkably enhanced, the overall performance of the composite filler is improved, the compatibility between the composite filler and polyurethane is enhanced by introducing a silane coupling agent, and the uniformity and stability of the membrane are ensured; the composite filler combines the impact resistance of the carbon nanotubes and the hardness and wear resistance of silicon dioxide, so that the film shows higher strength and durability under complex stress.
Owner:SHANGHAI ZITU NEW MATERIAL TECH CO LTD

Hydroxyapatite filler and preparation method thereof

The invention relates to a hydroxyapatite filler and a preparation method thereof, and relates to the field of implantation intervention biomedical materials. The preparation method comprises the following steps: dispersing hydroxyapatite in a solution, and adjusting the pH value to 4.5-6.5 to obtain a hydroxyapatite solution; dissolving oligomeric hyaluronic acid in the solution, adding EDC, and adjusting the pH value to 4.5-6.5 to obtain an oligomeric hyaluronic acid solution; mixing the hydroxyapatite solution and the oligomeric hyaluronic acid solution, and adding an accelerant to obtain a modified hydroxyapatite solution; centrifuging to obtain a first precipitate, dispersing the first precipitate in a mixed solution of ethanol and water, adding oligomeric hyaluronic acid, adjusting the pH value to 5.6-7.2, stirring at room temperature for a first time, and stirring at 2-8 DEG C for a second time; and centrifuging to obtain a second precipitate, re-dissolving the second precipitate, oscillating, and centrifuging again to obtain a third precipitate, namely the hydroxyapatite filler. Through the process, the oligomeric hyaluronic acid is effectively grafted on the hydroxyapatite, so that the hydroxyapatite filler is not easy to settle, and the pushing force is proper.
Owner:BAIHONG HEYI BIOTECHNOLOGY (SUZHOU) CO LTD

Thick gauge 800MPa hydroelectric steel with excellent core toughness and its manufacturing method

The invention discloses a thick specification 800MPa hydroelectric steel with excellent toughness in the core and a manufacturing method thereof. The components are C: 0.11-0.13%, Si: 0.1-0.15%, Mn: 1.05-1.15%, Cr: 0.4-0.5%, Ni: 0.55-0.65%, Mo: 0.35-0.40%, Cu: 0.35-0.40%, Nb: 0.01-0.03%, V: 0.055-0.065%, Ti: 0.006-0.012%, Als: 0.01-0.02%, B: 15-30ppm, Mg: 8-18ppm, N: 60-80ppm, and the rest is Fe. The steel plate is produced by continuous casting billet, and a 120-150mm thick steel plate is produced by controlling the heating temperature, rolling water cooling process, and quenching and tempering process, and the toughness at the center of the thickness is excellent.
Owner:MINMETALS YINGKOU MEDIUM PLATE

Ferronickel-based oxygen evolution electrode and preparation method thereof based on interface confinement

PendingCN121951581AImplement "directed construction"Implement a directed buildNickel compoundsElectrodesElectrolysisPhysical chemistry
The invention provides a ferronickel-based oxygen evolution electrode and a preparation method thereof based on interface confinement. The preparation method of the ferronickel-based oxygen evolution electrode comprises the following steps: preparing a precursor solution containing ferrous ions and ferric ions; the foamed nickel substrate is cleaned; the cleaned foamed nickel is tightly attached to the surface of the bottom of the reaction container, the foamed nickel is immersed in the precursor solution, under the room-temperature standing condition, divalent and trivalent iron ions are induced to be subjected to a synergistic hydrolysis and coprecipitation reaction through a physical interface confinement space formed by an attached interface, and the nickel-based composite material is obtained. Therefore, a high-activity and high-uniformity nickel-iron-based (oxygen) hydroxide catalyst layer (DIC-NiFeOOH / NF) is grown on the surface of the foamed nickel in situ. The method is simple in process, green and energy-saving, and the prepared nickel-iron-based oxygen evolution electrode shows excellent performance in oxygen evolution reaction and anion exchange membrane water electrolysis.
Owner:SUQIAN GREEN ENERGY HYDROGEN TECHNOLOGY CO LTD

A smart beamwidth optimization method based on reinforcement learning

The present invention discloses an intelligent beam width optimization method based on reinforcement learning. The method suppresses the beam drift effect in millimeter wave communication by dynamically adjusting the beam width. The algorithm models the beam width dynamic optimization problem as a Markov decision process, and selects the optimal beam width for data transmission at each decision moment. Its state design can characterize the severity of the beam drift effect of the current system from multiple angles, and each action corresponds to a different beam width, thereby optimizing and adjusting the width of the data transmission beam by sensing the speed of environmental changes. In each state, the optimal beam width for data transmission is selected according to the AC algorithm, and the policy network and value network are continuously updated during training to improve the rationality and reliability of model selection, while introducing a dynamic network update mechanism to reduce the computational burden of network updates. The throughput of the millimeter wave communication system under the beam drift effect is improved, thereby continuously ensuring the quality of the communication link.
Owner:SOUTHEAST UNIV

Optimized gradient compression method and device for enhancing distributed deep learning

InactiveCN118917384AImprove parallelismClosing the Performance GapBiological modelsHigh bandwidthAlgorithm
The invention discloses an optimized gradient compression method and device for enhancing distributed deep learning, and the method comprises the steps: building a communication threshold table through a cocktail system in a pre-training stage; in a formal training stage, a working node performs gradient calculation by using a local backup model to obtain a training gradient value; processing the training gradient value according to the communication threshold table to obtain a gradient value needing to be compressed, and sending the gradient value to a server node; and the server node processes the gradient value needing to be compressed to realize gradient compression. According to the method, a Cocktails system is introduced, on the basis of minimizing unnecessary compression operation, a compression task is transferred to a CPU to be executed, and the restriction of a communication task between nodes on calculation operation is relieved. In a high bandwidth environment, tensor calculation operations become sensitive to delays caused by compression operations on the GPU.
Owner:NAT UNIV OF DEFENSE TECH

A polydopamine modified modified nanographene and a preparation method and application thereof

ActiveCN121570605BRetains inherent absorptive propertiesImprove hydrophilicityCalcium fluxNeural cell
The application belongs to the technical field of biological medical materials, and discloses a polydopamine modified modified nano-graphyne and a preparation method and application thereof. In the method, the polydopamine coating is combined with the nano-graphyne for the first time to construct a GDY@PDA composite system, and the core defects of the blocky graphyne, such as easy agglomeration, poor water solubility and insufficient biocompatibility, are solved. The graphyne-based material is applied to the near-infrared two-region photothermal regulation of neural cell calcium influx for the first time, the calcium ion channel is activated through the local thermal effect of the material, and precise, remote and reversible calcium influx regulation is realized; the limitation that the traditional biocompatibility modification is easy to fall off and difficult to functionalize is broken through, the polydopamine modification makes the material have both photothermal efficiency and biological safety, and the application of the graphyne in the treatment of neural diseases (such as neural cell injury repair) is expanded.
Owner:BEIJING INSTITUTE OF PETROCHEMICAL TECHNOLOGY

A synthetic method for multidimensional transformation of ZIF-8 and ZIF-L

The invention discloses a synthesis method for multidimensional conversion of ZIF-8 and ZIF-L, by adopting aqueous phase to synthesize ZIF-8 crystals, and using hexadecyltrimethylammonium bromide (CTAB) as a capping agent to control the transformation of ZIF-8 and ZIF-L crystalline phases and the morphology and size of grains. By controlling the addition amount of CTAB, selective isotropic and anisotropic nanoscale growth is achieved, and one-dimensional rod-shaped, two-dimensional sheet-shaped and three-dimensional nano-flower-shaped ZIF-8 crystals are successfully prepared. Wherein the two-dimensional sheet structure is ZIF-L crystalline phase, which belongs to the metastable crystalline phase of ZIF-8 crystal, and has a trend of topological phase transition from two-dimensional layered structure to three-dimensional framework material. The regulation of microstructure is extremely important for studying the structural change law of ZIFs and expanding the application field.
Owner:BEIJING INFORMATION SCI & TECH UNIV

Ground-based space object monitoring radar atmospheric refraction error correction method

The application discloses a ground-based space target monitoring radar atmospheric refraction error correction method. The ground-based space target monitoring radar atmospheric refraction error correction method comprises the following steps: based on the vertical distribution of atmospheric refraction index and the radar elevation angle measurement value, the signal propagation path is accurately recursively calculated by using the refraction theorem and geometric relationship, the corresponding elevation angle, geocentric angle and target apparent distance of different heights are calculated, and a one-to-one corresponding relationship between the apparent distance and the distance, the elevation angle measurement error is established; then, according to the radar distance measurement value, the radar distance and the elevation angle measurement error are calculated by interpolation, and finally, the distance and the elevation angle measurement error caused by the atmospheric refraction of the space target monitoring radar are corrected. The ground-based space target monitoring radar atmospheric refraction error correction method has higher error calculation precision under the condition of given layer interval, has more stable error correction performance, and can more accurately correct the atmospheric refraction error of the space target monitoring radar.
Owner:NAT UNIV OF DEFENSE TECH

Polymer / inorganic heterostructure composite film and preparation method and application thereof

The application belongs to the field of organic-inorganic heterostructure thin films, and relates to a polymer / inorganic heterostructure composite thin film and a preparation method and application thereof. The composite thin film is formed by compounding PEDOT:PSS and ITO nanocrystals, PEDOT:PSS forms a flexible network skeleton, the growth orientation of ITO nanocrystals is a (222) crystal face, a (400) crystal face or a (211) crystal face, and the ITO nanocrystals are uniformly dispersed and bonded in the PEDOT:PSS network. The flexible organic polymer PEDOT:PSS and the brittle ceramic material ITO are compounded, the surface state and the work function of the two-phase materials are regulated, the heterostructure composite thin film with a molecular-level interface combination is constructed, and the critical mechanical properties, the interface toughness and the electron mobility of the thin film are simultaneously improved, so that the demand of the flexible electronic device for the high-performance and high-reliability thin film material is met.
Owner:QILU UNIVERSITY OF TECHNOLOGY (SHANDONG ACADEMY OF SCIENCES)

Semi-supervised remote sensing image change detection method and system based on neighborhood relation propagation

ActiveCN122289944BImprove Spatial Consistencyclear boundaries
This invention discloses a semi-supervised remote sensing image change detection method and system based on neighborhood relationship propagation, belonging to the field of computer vision and remote sensing image processing technology. It includes: acquiring dual-temporal remote sensing image pairs and dividing them into labeled and unlabeled datasets; constructing a teacher-student framework semi-supervised model, calculating the squared Euclidean distance between the center pixel and its neighboring pixels, converting it into a relevance weight matrix using Softmax, and performing neighborhood-weighted aggregation on the initial prediction map to obtain a corrected prediction map; calculating the supervised loss using labeled samples, and calculating the unsupervised loss using unlabeled samples through weak perturbation to generate pseudo-labels and strong perturbation for consistency learning; jointly optimizing the student model parameters and updating the teacher model using exponential moving average; inputting the dual-temporal images into the trained student model, outputting a probability map, and generating a binary change detection mask. This invention enhances spatial consistency through neighborhood relationship propagation, improving detection performance when labels are limited.
Owner:江苏源驶科技有限公司

Production method of mixed alcohol amine without rectification and ammonia recovery

The invention discloses a production method of mixed alcohol amine without rectification and ammonia recovery. The production method specifically comprises the following steps: S1, a reaction stage of ethylene oxide and ammonia water; s2, a further reaction stage with epoxypropane; the invention relates to the technical field of chemical synthesis. According to the production method of the mixed alcohol amine without rectification and ammonia recovery, a method for producing ethanolamine and isopropanolamine without distillation and free ammonia recovery is adopted, the links of distillation and free amine recovery are omitted, the cost is greatly saved, the method is simple and convenient, a water absorption molecular sieve is used as a catalyst, and the production cost is greatly reduced. According to the method, the catalytic water consumption is greatly reduced, so that the energy consumption is reduced, liquid ammonia is directly used as a raw material, the material cost and the operation cost are reduced, meanwhile, the reaction rate is increased, the generation of impurities is reduced, the decomposition of the product is avoided, the purity of isopropanolamine is increased (99% or above), and the method is suitable for large-scale production and has a wide application prospect. And the production process flow is simplified.
Owner:JIANGSU RENAI CONSTR TECH DEV CO LTD

Ejector synergistic CO2 heat pump air conditioning system for new energy automobile and control method

The invention discloses an ejector synergistic CO2 heat pump air-conditioning system for a new energy automobile. The ejector synergistic CO2 heat pump air-conditioning system comprises an air-conditioning module and an air-conditioning box module. The air conditioner module is composed of a compressor, an automobile outer side draught fan, an automobile outer heat exchanger, an automobile inner main heat exchanger, an automobile inner auxiliary heat exchanger, a heat regenerator, a gas-liquid separator, an ejector, an electronic expansion valve, a first four-way reversing valve, a second four-way reversing valve and a three-way valve. The air conditioner box module is composed of an air conditioner box body, an air conditioner air door and an air conditioner draught fan. In addition, the invention also discloses six working modes and a control method of the system. According to the system, the ejector is introduced to recycle expansion work in the refrigerant throttling process, the high-temperature refrigeration performance and the low-temperature heating performance of the CO2 heat pump air conditioner can be effectively improved, and the system is simple in structural design and stable and reliable in operation.
Owner:ZHENGZHOU UNIV

Neural network model-based few-sample named entity identification method and system

PendingCN121997927AImprove inference speedMeet the requirements for rapid adaptationSemantic analysisBiological modelsNamed-entity recognitionText entry
The invention discloses a few-sample named entity recognition method and system based on a neural network model, and the method comprises the following steps: obtaining a to-be-recognized original input text, and retrieving from an external knowledge base to obtain a group of candidate knowledge statements; a dynamic similarity threshold algorithm is adopted to filter the candidate knowledge statements, and enhanced knowledge is screened out; inputting into a first encoder, and generating a context text representation matrix of deep fusion knowledge; inputting the matrix into a sequence labeling classification layer, classifying each lexical element, and identifying a boundary span of a potential entity in a text; and inputting a text corresponding to the identified entity boundary span into a second encoder to generate an entity embedding representation, and determining a final category of the entity by calculating a distance between the entity embedding representation and a predefined category prototype to complete named entity identification. According to the method, the problems caused by data sparseness and semantic fuzziness in a few-sample scene are solved, and the accuracy of named entity recognition is improved.
Owner:GUANGDONG UNIVERSITY OF FOREIGN STUDIES

Unmanned aerial vehicle hovering point and flight path joint optimization scheme for data collection performance improvement

PendingCN121832565AExcellent data recovery accuracyEfficient and stable deploymentVehicle position/course/altitude controlPosition/direction controlUncrewed vehicleData acquisition
The invention discloses an unmanned aerial vehicle hovering point and flight path joint optimization scheme (DGQDN) for data collection performance improvement. The DGQDN comprehensively optimizes the hovering point and the flight path of the unmanned aerial vehicle, and aims to maximize the system utility. According to the scheme, the system utility not only covers the total amount of sensor network data acquisition, but also considers optimization of the number of hovering points and the repeated coverage rate of the sensor nodes. In the scheme, after the unmanned aerial vehicle completes data acquisition, the matrix filling technology is adopted to accurately recover the sensing data of an unacquired area. The DGQDN starts from three dimensions of hovering point optimization, flight path planning and data recovery and expansion, the data acquisition efficiency of the unmanned aerial vehicle is systematically improved, and an innovative solution is designed for each dimension. A large number of simulation experiments prove that the DGQDN is obviously superior to an existing scheme in different data acquisition scenes, and excellent performance and wide applicability of the DGQDN are shown.
Owner:GUANGXI UNIV

Paper poking piece separating structure of chain type transmission paper feeding equipment

ActiveCN224212034USolve the problem of electrostatic adsorptionImprove separation efficiencyArticle separationProcess engineeringIndustrial engineering
The utility model relates to the technical field of packaging paper feeding and cutting, and discloses a paper stirring piece separating structure of chain type transmission paper feeding equipment, which comprises a separating plate fixed on a rack, a plurality of air outlet grooves are uniformly arranged at the upper part of the separating plate, a communicating pipe is arranged on one side of each air outlet groove, the communicating pipe is connected with an air flow cleaner, and the air flow cleaner is connected with an air inlet pipe of the chain type transmission paper feeding equipment. And a stripping plate is arranged at the upper part of the communicating pipe and is used for stripping the adsorbed paper. According to the utility model, a'mechanical + pneumatic 'double-acting separation mode is formed through sequential cooperation of stripping elastic sheet mechanical triggering and high-pressure airflow, the problem of electrostatic adsorption of multiple pieces of paper is effectively solved, the separation efficiency is improved by more than 40%, the airflow cleaner directionally sprays compressed air through the air outlet groove, static electricity on the surface of the paper is neutralized while a physical isolation layer is formed, and the separation efficiency is improved. And the electrostatic adsorption force is reduced to 0.5 N or below, so that secondary adhesion is fundamentally avoided.
Owner:SAGA COMPUTER NUMERICAL CONTROL CO LTD

Quantitative diagnosis method for freshwater fish diseases based on openCV weak supervision and resnet50-adca double branch

PendingCN122551037Alow costSolve the pain points of traditional manual annotation
A quantitative diagnosis method for freshwater fish diseases based on OpenCV weak supervision and ResNet50-ADCA dual-branch model is proposed. The method consists of the following steps: (1) Image acquisition and preprocessing: Based on the publicly available freshwater fish disease dataset, a balanced Salmon++ dataset is constructed by combining samples collected from aquaculture sites; (2) Weakly supervised lesion segmentation and labeling: (2.1) HSV color space targeted segmentation; (2.2) Iterative optimization mechanism; (2.3) Disease grading standard; (3) ResNet50-ADCA dual-branch model construction: (3.1) Backbone network and transfer learning; (3.2) SEA channel attention embedding; (3.3) Dual-branch multi-task output layer design; (3.4) Classification branch; (3.5) Grading branch; (3.6) Loss function design; (4) Model training and optimization. This invention balances labeling efficiency, diagnostic accuracy, and quantitative grading capability, and can effectively solve the problems of high labeling cost, poor adaptability to complex environments, lack of quantitative evaluation function, and imbalance between accuracy and speed in the existing technology.
Owner:JIANGSU OCEAN UNIV

Integrated energy system state estimation method based on LSTM automatic encoder

The invention relates to the technical field of power systems, and particularly discloses an integrated energy system state estimation method based on an LSTM automatic encoder, and the method comprises the following steps: building an IES state estimation model, and describing an energy coupling relation in the IES through a constraint equation; state estimation of the IES is carried out based on a weighted least square method; constructing an LSTM automatic encoder, and training the LSTM automatic encoder by using historical operation data of the IES as a training data set to obtain a trained LSTM automatic encoder model; inputting to-be-estimated IES system time sequence data into the trained LSTM automatic encoder model, and obtaining a state estimation result of the integrated energy system through forward propagation of the model; determining a loss function, wherein the loss function comprises prediction loss and regularization loss; the method has the advantages that the long-term dependency relationship and the nonlinear characteristics in the time sequence data can be effectively learned and captured, so that the accuracy and the reliability of state estimation are improved.
Owner:SICHUAN UNIV

A gas-liquid buffer

InactiveCN111396490BAccurate resetUniform impact forceSpringsGas and liquid based dampers
The present invention provides a gas-liquid buffer, which belongs to the technical field of buffer equipment, and comprises an outer cylinder body and an inner cylinder body; one end of the outer cylinder body is an open end, and a guide rod is arranged inside, and a guide circle is fixed to the end of the guide rod close to the open end of the outer cylinder body; one end of the inner cylinder body is inserted into the outer cylinder body from the open end of the outer cylinder body, and the other end is located outside the outer cylinder body; a guide plate is fixed to the insertion end of the inner cylinder body, and the guide plate is provided with a guide hole; the inner diameter of the guide hole is larger than the outer diameter of the guide rod and smaller than the outer diameter of the guide platform; the inner cylinder body is sleeved on the guide platform and the guide rod through the guide plate, and an isolating piston that moves along its axial direction is also arranged in the inner cylinder body. The gas-liquid buffer provided by the present invention can avoid the situation that the traditional buffer is prone to being unable to be completely reset and misaligned, and has the advantages of large energy absorption, long service life, simple structure and automatic and rapid reset.
Owner:LIAONING QINGYUAN FIRST BUMPER MFG

Batch user migration management method and device in weak network environment and storage medium

The present application relates to weak network environment batch user migration management method, device and storage medium, read operation task data from local offline file to execute operation task based on task data; Group operation tasks according to the network address of baseboard management controller, adopt asynchronous I / O concurrent processing mechanism, and execute operation tasks of different network addresses of baseboard management controller concurrently, the present application realizes concurrent operation to user, improves operation efficiency; In the same baseboard management controller, operation tasks are executed in turn according to the definition order of local offline file to avoid concurrent conflict; The business logic of all operation tasks is concentrated in the core module, and the calling module is established as the program entry and dispatcher to call the core module to execute operation tasks, so that the weak network environment batch user migration management is completed. The present application can improve operation efficiency, especially in weak network environment, the performance advantage of the present application is more significant compared with the traditional scheme.
Owner:POWERLEADER COMPUTER SYST CO LTD

A multi-scale energy phase fusion and adaptive difference domain-invariant underwater acoustic feature extraction method

PendingCN122734497AResistant to amplitude fadingObvious performance advantage
The application discloses a multi-scale energy-phase fusion and adaptive difference domain-invariant underwater acoustic feature extraction method, and particularly relates to the design of four core stages of multi-time-frequency resolution spectrum extraction, multi-scale energy-phase fusion, learnable feature alignment and channel splicing, and adaptive time-frequency difference reconstruction. In view of the performance decline problem of the existing method in cross-sea area deployment, the method has the functions of underwater acoustic target robust recognition and generalization application of resisting amplitude fading and suppressing stationary environment artifacts as a light-weight front-end feature extraction module.
Owner:HARBIN ENG UNIV

Digital forest compass and measuring system

The utility model discloses a digital forest compass, and belongs to the technical field of forest resource investigation. Comprising a measuring device integrated with a single-point ranging radar, an attitude sensor and an industrial camera; and the tablet computer is connected with the measuring device and is used for processing, recording and storing the data extracted by the measuring device. The utility model also discloses a digital forest compass measurement system, which comprises measurement hardware which is the digital forest compass; and the measuring software consists of embedded software and PC (Personal Computer) end software. The whole device is small in size, convenient to carry and capable of being operated by one person; compared with a traditional measuring instrument, data does not need to be manually recorded, input and corrected, real-time images and virtual sight beads presented by the industrial camera on software replace a sighting telescope of a traditional compass for target alignment, and efficiency is greatly improved. In an efficiency measurement test, the measurement efficiency of the instrument is twice that of a traditional instrument.
Owner:ZHEJIANG FORESTRY UNIVERSITY +2

A composite catalyst with high resistance to coking in methane dry gas reforming and its application

ActiveCN117983278BImprove anti-coking performanceExtended service lifeCatalyst carriersHydrogen
This invention discloses a composite catalyst with high resistance to coking in methane dry gas reforming. The composite catalyst is composed of three components: a supported nickel-based catalyst, a nitrogen-doped carbon-supported molybdenum catalyst, and an inert support. The invention also discloses its application in methane dry gas reforming. After hydrogen reduction, the composite catalyst exhibits excellent activity and stability under high space velocity reaction conditions in methane dry gas reforming, overcoming the deficiency of Ni-based catalysts in methane dry gas reforming that are prone to coking and deactivation under high-temperature conditions.
Owner:ZHEJIANG NORMAL UNIV