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

129results about How to "Efficient aggregation" patented technology

Multi-modal emotion understanding method based on cross-modal semantic alignment and interactive learning

The invention discloses a multi-modal emotion understanding method based on cross-modal semantic alignment and interactive learning. The method comprises the following steps: acquiring multi-modal initial data; performing semantic quantification processing on the initial data of each mode; according to the semantic feature set, text modal dominant cross-modal feature aggregation learning is carried out; according to the multi-modal semantic aggregation feature expression, performing intra-modal semantic relationship learning of the multi-modal features, determining a context relationship between intra-modal semantics Tokens, and performing semantic alignment with a text modal to enhance semantic consistency of different modal feature expressions; carrying out fusion processing on the multi-modal features by adopting a multi-modal fusion strategy to obtain fusion features; and establishing an emotion understanding model according to the fused features, and realizing multi-modal emotion understanding processing according to the emotion understanding model. The embodiment of the invention can improve the sentiment analysis performance, is beneficial to promoting effective fusion of different modes, and can be widely applied to the technical field of computers.
Owner:SUN YAT SEN UNIVERSITY SHENZHEN +1

Customer service data aggregation method based on ODPS and server

The invention discloses an ODPS-based customer service data aggregation method and server, and relates to the field of electric digital data processing, and the method comprises the steps: carrying out the field classification of collected source data, and obtaining a plurality of source data sets; loading the source data sets to an ODPS platform, and performing first aggregation processing on the source data in the plurality of source data sets by using the ODPS platform to obtain a plurality of field wide tables; pulling the field wide table to a DRDS platform, and performing distributed data processing on the field wide table by using the DRDS platform to obtain a plurality of pieces of distributed data; determining a target data feature corresponding to a target report in a report request instruction according to the report request instruction sent by the user terminal; and extracting multiple pieces of target data corresponding to the target data features in the multiple pieces of distributed data, and performing second aggregation processing on the target data to generate a target report. By implementing the method, the data processing efficiency can be improved when the multi-azimuth data display of the customer service staff is carried out.
Owner:SHANGHAI SHENXUE SUPPLY CHAIN MANAGEMENT CO LTD

Deep learning-based tomato maturity detection system and method

The invention discloses a tomato maturity detection system and method based on deep learning, and relates to the technical field of fruit and vegetable detection models. Comprising the following steps: acquiring data from a robot perspective based on a natural scene of a tomato plantation, and making a tomato maturity detection data set; on the basis of a YOLO v5s model, an improved model YOLOv5s BCD model fusing a CA attention mechanism, a BiFPN neck feature fusion network and a Decup led Head head decoupling network is provided and serves as a tomato maturity detection model of the invention; experimental research of model training and effect comparison is introduced, the conditions of model evaluation and optimization are performed through a model comparison test, the tomato maturity grade is judged on the basis of the fruit red surface coloring proportion and in combination with fruit appearance characteristics, tomatoes with the required maturity grade are collected, the light weight and accuracy of edge side deployment are adapted, and the tomato maturity grade evaluation method is suitable for the tomato maturity grade evaluation and the tomato maturity grade evaluation and the tomato maturity grade evaluation and the tomato maturity grade evaluation and the tomato maturity grade evaluation. The system can be used as a visual system of a tomato harvesting robot, and technical support is provided for automatic picking.
Owner:BENGBU COLLEGE

Vehicle driving track prediction system and method based on improved graph neural network model

The invention belongs to the field of traffic control systems, and relates to a vehicle driving track prediction system and method based on an improved graph neural network model, and the system comprises an FPGCN model and a Social LSTM model. The FPGCN model comprises a graph convolution layer, a Dropout layer, a differential pooling layer, a feature fusion module and a full connection layer; obtaining node feature vectors from graph structure data through a graph convolution layer, a Dropout layer and a differentiable pooling layer, then dividing and conquering node features, generating global features through a dense attention mechanism and principal component analysis, and generating local features through virtual filling and matrix calculation; after global features and local features are obtained, a feature fusion module is used for fusion, then the features are input into a full connection layer, and the influence score of a historical trajectory diagram on a current state diagram is calculated; integrating the preprocessed driving data and the data of which the score is calculated by the FPGCN model, and inputting the integrated data into a Social LSTM model; according to the system, while the model expression ability is maintained, the parameter quantity is remarkably reduced, the calculation logic is optimized, and the prediction accuracy is improved.
Owner:CHANGCHUN UNIV

Continuous production process of CO2-based multipolymer

The invention discloses a continuous production process of a CO2-based multipolymer, and belongs to the technical field of polycarbonate synthesis. The technological process comprises the following steps: firstly, preparing an auxiliary agent, a solid-state monomer, a liquid-state reaction monomer and any solvent into a mixed solution in a premixer (1); continuously feeding carbon dioxide and the mixed solution into a multi-stage continuous reactor for mixing and heating to initiate a reaction, wherein the total reaction residence time of the materials in the multi-stage continuous reactor is controlled to be 10 minutes to 5 hours; and after the reaction is completed, carrying out post-treatment to complete the continuous production of the CO2-based multipolymer. The process particularly optimizes the effective continuous polymerization of high-molecular and high-viscosity materials of the CO2-based multipolymer. Through accurate control of the multi-stage tubular reactor, appropriate reaction viscosity is realized, further reaction uniformity of materials in the screw extrusion reactor is ensured, and reaction efficiency is improved.
Owner:SHANDONG LECSIN GREEN TECH CO LTD

Unmanned aerial vehicle video target detection method and system based on space-time correlation

The invention discloses an unmanned aerial vehicle video target detection method and system based on space-time correlation, and the method comprises the steps: S01, inputting the video image data of an unmanned aerial vehicle into a feature extraction network, and obtaining an initial feature sequence; s02, embedding the initial feature sequence into a spatial position, performing deep convolution processing, performing space-time modeling by using a state-space equation, and fusing with the proportionally adjusted initial feature through residual connection to form a space-time feature; s03, inputting the spatial-temporal features into an attention network with residual connection, setting different weights according to appearance similarity, and fusing the spatial-temporal features through the residual connection; s04, performing feature fusion on the initial features and the spatio-temporal features after foreground attention processing; and S05, inputting the space-time fusion feature to a detection head to obtain a target detection result and outputting the target detection result. The method has the advantages of being simple in implementation method, few in needed computing resources, high in real-time performance and detection precision, high in expandability and the like.
Owner:NAT UNIV OF DEFENSE TECH

AI identification method and system for millimeter wave image overlapping target, and model training method

The invention discloses an AI identification method and system for a millimeter wave image overlapping target based on a three-branch network, and a model training method, and the method comprises the steps: carrying out the cooperative work of three independent branch networks: an analysis branch network, an aggregation branch network, and a prediction boundary branch network; the problem of security inspection target detection in millimeter wave images with low resolution, mutual interference or mutual overlapping is solved. The analysis branch network stores detail information in the high-resolution feature map; the aggregation branch network realizes feature position offset alignment of high-resolution and low-resolution feature maps through a recursion fusion module; and the prediction boundary branch network extracts high-frequency features, and enhances perception and prediction of the target edge. The multi-branch network structure effectively realizes image segmentation and identification through fusion among branches. While the segmentation performance of the model is improved, the model significantly reduces the calculation complexity, reduces the detection error and missed detection probability in the millimeter wave image, and finally achieves the precise detection of overlapped objects.
Owner:UNIV OF SHANGHAI FOR SCI & TECH

Internet of vehicles federated learning method based on active learning and semi-supervised learning

The invention discloses an Internet of Vehicles federated learning method based on active learning and semi-supervised learning, and belongs to the field of Internet of Vehicles federated learning, and the method comprises the steps: training a global model at a cloud server side based on mark data, and updating the parameters of the global model; through a plurality of road side units, the global model parameters are distributed to automatic driving automobiles in signal coverage ranges of the road side units; on the basis of the unmarked data of the autonomous vehicle, updating the parameters of the vehicle side model according to the global model parameters and the unmarked data; parameters of the vehicle side model of each automatic driving vehicle are sent to a cloud server side through a road side unit; and randomly determining an aggregation client set, aggregating the parameters of the autonomous vehicle side models in the aggregation client set at the cloud server side to obtain global model parameters, and copying the global model to a teacher-student network of the cloud server side until a communication round is reached. The problem that the effectiveness is too low when traditional federal learning is applied to the Internet of Vehicles is solved.
Owner:BEIJING JIAOTONG UNIV

Efficient oil removal and purification equipment in deep sea gas field ethylene glycol enrichment system

The invention discloses efficient oil removal and purification equipment in a deep sea gas field ethylene glycol enriching system, and belongs to the technical field of petroleum treatment, the efficient oil removal and purification equipment comprises purification equipment composed of multiple stages of oil removal devices, the oil removal devices are connected stage by stage through communicating pipes, and each oil removal device comprises a separation pipe, a coalescence module and an oil removal film module, and the coalescence module and the oil removal film module are arranged in the separation pipe; the coalescence module and the oil removal film module are arranged, a coalescence technology and an oil removal film technology are combined, a set of dual separation system is constructed, the combined technology shows obvious advantages when an oil-water mixture with high emulsification degree is treated, a super-hydrophilic fiber bed layer is arranged in the oil removal film module, the fiber bed layer utilizes a water film barrier effect, and the oil removal effect of the oil-water mixture is improved. According to the oil-water separator, the oil-water mixture can be further captured and polymerized without being completely polymerized by the coalescer, and a multi-layer design is adopted, so that the oil-water mixture is treated for multiple times when passing through the equipment, the whole-process treatment from crude separation to fine separation is realized, and the overall separation efficiency of the equipment is greatly improved.
Owner:NANJING PINMAO NEW ENERGY TECH CO LTD

Double-bridge normal form U-shaped medical image segmentation network

The invention discloses a double-bridge normal type U-shaped medical image segmentation network, which comprises an encoder, a double-bridge framework and a decoder, the encoder adopts a backbone network, the double-bridge architecture comprises a multi-scale bridge and a single-scale bridge, the multi-scale bridge adopts an SCAM, the single-scale bridge adopts a CAM, and the decoder adopts an RMSCAM; the encoder extracts multi-stage image features from the medical image of the single-channel gray scale, and the features extracted by the encoder in four stages are fed into the double-bridge architecture; the multi-scale bridge receives features extracted from the four stages of the encoder and captures global features; the single-scale bridge receives features extracted by the encoder in a single stage and captures local features; the global features and the local features are added and fused and then sent to a multi-stage decoder; the decoder restores the feature map to obtain a final prediction map; according to the method, the problems of semantic difference between different stage features extracted by the encoder and semantic difference between the encoder and the decoder can be effectively solved.
Owner:DALIAN NATIONALITIES UNIVERSITY

Generalized rapid Mesh extraction method based on Gaussian

The invention discloses a generalized rapid Mesh extraction method based on Gaussian, and the method comprises the steps: configuring a training data collection system, and carrying out the marking and preprocessing of data; the method comprises the following steps: constructing a generalized rapid Mesh extraction network based on Gaussian, and training the rapid Mesh extraction network; a training model of a generalized rapid Mesh extraction network based on Gaussian is used, testing is carried out under multiple data sets and different scenes, index evaluation is carried out on a result, the performance of the Mesh extraction method based on the Gaussian model can be further improved through a self-adaptive adjustment strategy and an optimized parameter estimation method, and the performance of the Mesh extraction method based on the Gaussian model is improved. The method has important research significance in theory, particularly has wide practical application prospects in scenes needing to process large-scale and high-dimensional complex data, aims to overcome the defects in the prior art and improve the adaptability and efficiency in different data sets and application scenes, and provides an innovative solution for the field of three-dimensional data processing. And further development and application expansion of the Mesh extraction technology are promoted.
Owner:HUBEI CHUCK TECH CO LTD

Smoke image detection method based on split Top-K attention mechanism

The invention relates to the technical field of fire detection, in particular to a smoke image detection method based on a split Top-K attention mechanism, and the method comprises the steps: training a smoke network model through employing a full supervision method; an extraction module of a third stage and an extraction module of a fourth stage of a feature extraction network of the smoke network model are constructed based on a Transform module of a split Top-K attention mechanism, global features can be effectively extracted from long-distance noise of an image, local and global features are effectively aggregated, and meanwhile noise of an irrelevant long-range dependency relationship is reduced. And the cross-stage fusion module aggregates the output from the four stages, so that the flexibility and fitness of smoke characteristics on different scales can be improved, and the detection precision is improved. And inputting an aggregation result into a linear classifier to predict scores of four preset categories. In conclusion, the accuracy of smoke image category detection under the foggy weather condition is improved.
Owner:SUN YAT SEN UNIV

Method for preparing composite solid electrolyte through Lewis fluorine supply nano additive

The invention provides a method for preparing a composite solid electrolyte through a Lewis fluorine supply nano additive, and the composite solid electrolyte comprises the following components in parts by weight: 10-20 parts of a polymeric monomer, 2-6 parts of a lithium salt, 1-4 parts of a nano functional additive and 5-12 parts of an organic solvent. When the prepared composite solid electrolyte is used for a lithium ion battery, the composite solid electrolyte has the characteristics of no need of an additional initiator, uniform polymerization reaction, high ionic conductivity and the like, and the stability, the rate and the cycle performance of the semi-solid lithium ion battery can be improved; according to the preparation method, no extra initiator is added into the preparation raw materials of the in-situ cured solid electrolyte, the lithium salt is instably decomposed into the lewis acid in a heating state, the lewis acid is utilized to efficiently initiate ring-opening of cyclic ether to perform polymerization reaction, no initiator is left, and no extra side reaction is generated to cause reduction of battery performance.
Owner:SHENZHEN GUKE NEW ENERGY TECHNOLOGY CO LTD

Virtual power plant distributed resource aggregation optimization method and device under homomorphic encryption

The invention provides a method and equipment for aggregation optimization of distributed resources of a virtual power plant under homomorphic encryption. The method comprises the steps of obtaining distributed resource data of a virtual power plant; performing classification; homomorphic encryption processing is carried out to obtain encrypted data; performing homomorphic encryption optimization to obtain a ciphertext of the distributed resource data of the virtual power plant; based on the encrypted virtual power plant distributed resource data, determining a target and constraint conditions of virtual power plant distributed resource aggregation optimization, and constructing a virtual power plant distributed resource aggregation optimization model; and solving the virtual power plant distributed resource aggregation optimization model through a heuristic algorithm to obtain a virtual power plant distributed resource aggregation optimization scheme. Under the condition that original data are not exposed, efficient aggregation and optimization of the distributed resources of the virtual power plant are achieved. The security and privacy of the distributed resource data of the virtual power plant are ensured, the energy utilization efficiency and the system stability are improved, and technical support is provided for the construction of an intelligent power grid and the virtual power plant.
Owner:BEIJING CHINA POWER INFORMATION TECH

Antistatic heat-insulating coating and preparation method thereof

The invention provides an antistatic heat-insulating coating and a preparation method thereof, and relates to the technical field of coatings. The antistatic heat-insulating coating is prepared from polyacrylate, titanium dioxide, antistatic heat-insulating filler, hydrophilic aerogel, diatomite, water and other raw materials, the polyacrylate comprises Vitrimer xylose polyacrylate with a specific structure, the Vitrimer xylose polyacrylate is used as a substrate material of the coating, and the mechanical property, heat resistance, durability and the like of the coating can be remarkably improved; furthermore, the antistatic heat-insulating filler comprises an antistatic heat-insulating composite material which is formed by compounding a light material, polypyrrole and a carbon material, the composite material is prepared by an in-situ polymerization method, and a unique structure that a composite microlayer is formed by compounding the polypyrrole and the carbon material on the surface of the light material is formed; the anti-static heat-insulating composite material and other raw materials in the coating cooperate with each other to act together, so that the prepared coating has good anti-static performance and heat-insulating performance.
Owner:DALIAN UNIV +1

Visual text interaction-oriented multi-modal data fusion method and system

The invention discloses a multi-modal data fusion method and system for visual text interaction, and the method comprises the steps: carrying out the preprocessing of original video data and original text data; performing visual feature embedding on the original video data based on a space-time diagram convolution model to obtain visual feature representation; performing hierarchical semantic aggregation processing on the original text data based on a BERT model to obtain text feature representation; generating an aggregated appearance node representation and an aggregated motion node representation using the visual feature representation and the textual feature representation based on the interaction relationship matrix to generate a target visual representation; and fusing the target visual representation and the text feature representation to obtain a visual text joint representation, and executing a visual text interaction task based on the visual text joint representation. According to the method, the problems of shallow semantic fusion level and simple visual text interaction in the existing multi-modal data fusion technology are solved.
Owner:SUN YAT SEN UNIV +1

Vapor putty for special vehicle body material and preparation method thereof

The invention discloses putty for a special vehicle body material and a preparation method of the putty, and relates to the technical field of putty. The putty for the special vehicle body material is prepared from modified unsaturated polyester resin, a dispersing agent, modified talcum powder, bentonite, fumed silica, titanium dioxide, cobalt iso-octoate, cyclohexanone peroxide and dibutyl phthalate. The modified unsaturated polyester resin is obtained by reacting a modified unsaturated polyester resin precursor with hydroxyalkyl-terminated modified polysiloxane and then reacting with isocyanate terpyridyl; the modified talcum powder is obtained by reacting talcum powder modified by p-aminophenyltrimethoxysilane with chlorobenzene, cyclizing by using powdered sulfur, reacting with 1-iodoheptane, formylating by using phosphorus oxychloride, and grafting p-trifluoromethylaniline and arginine. The putty for the special vehicle body material prepared by the invention has the advantages of quick drying, antibacterial property and heat resistance.
Owner:WENGYUAN COUNTY ZHONGHAN MINFU COATING CO LTD

Novel compound and pharmaceutical composition comprising same as active ingredient

The present invention relates to a novel compound which modulates the function and differentiation of a cytoskeleton involved in pulmonary fibrosis and thus can be used for preparing a medicament capable of treating related diseases. It has been determined that: the novel compound according to the present invention or a pharmaceutically acceptable salt thereof has an effect of modulating actin polymerization required for cytoskeleton formation; and the expression of alpha-SMA and F-actin, which are main factors causing fibrosis, is effectively inhibited in human lung tissue-derived fibroblasts (MRC5). Furthermore, it has been determined that when activated macrophages are treated with the compounds, the level of inflammation of inflammatory cytokines caused by the macrophages is reduced. Therefore, the novel compound according to the present invention is expected to be effectively used in the prevention and treatment of diseases associated with pulmonary fibrosis as a preparation having an effect of inhibiting or reducing pulmonary fibrosis.
Owner:SAPIENS BIOTECHNOLOGY CO LTD

Cross-domain transfer learning laser ultrasonic layering defect detection method suitable for composite material and application of cross-domain transfer learning laser ultrasonic layering defect detection method

The invention provides a cross-domain transfer learning laser ultrasonic layering defect detection method suitable for a composite material and application thereof.The method comprises the steps that firstly, multi-channel input containing a frequency domain, a time-frequency domain and Koopman operator characteristics is constructed for the problem of data distribution deviation caused by the batch difference of the composite material; secondly, extracting and fusing high-dimensional spatial-temporal features through a multi-channel attention mechanism and a graph convolutional network; and finally, migrating the defect features of the source domain to the target domain in combination with dynamic distribution self-adaption and domain adversarial training strategies. The method can effectively overcome model performance reduction caused by material attribute change, realizes high-precision layering defect identification and imaging under the condition of no target domain label, and is suitable for composite material detection in a complex industrial scene.
Owner:ZHEJIANG UNIV

Image super-resolution method based on layered sensing distillation

The invention discloses an image super-resolution method based on hierarchical perception distillation. The method comprises the following operation steps: step 1, inputting a low-resolution image data set; 2, a shallow feature extraction process; 3, a deep feature extraction and enhancement process: processing the image through a deeper network layer to extract complex high-level features; step 4, image reconstruction and optimization: after feature extraction and enhancement, the model integrates the extracted features; and 5, outputting a reconstruction result: outputting the processed and reconstructed image as a high-resolution version, wherein the output result shows significant visual improvement. Compared with the prior art, the method has the advantages that the network structure of the super-resolution reconstruction model is optimized, and the advantages of the convolutional neural network and the visual transformer are combined, so that the model which is lighter and more efficient is developed.
Owner:XINJIANG UNIVERSITY

Transparent photosensitive resin for DLP 3D printing and preparation method thereof

The invention discloses transparent photosensitive resin for DLP 3D printing and a preparation method of the transparent photosensitive resin. The transparent photosensitive resin for DLP 3D printing comprises the following components in percentage by weight: 40%-60% of tetrahydrofuran acrylate, 20%-30% of bifunctional polyurethane acrylate, 19.5%-29.5% of a reactive diluent, 0.02%-0.03% of a light absorber, 0.2%-0.3% of a photoinitiator Irgacure-2959, 0.1%-0.3% of a defoaming agent and 0.1%-0.3% of a flatting agent, and the transparent photosensitive resin for DLP 3D printing is prepared by fully stirring and mixing the components at room temperature. The photosensitive resin for DLP 3D printing is transparent, low in viscosity and high in product speed, and the rapid formability of a 3D printing product is guaranteed; after curing, the volume shrinkage change is small, and the dimensional stability of the product is high; the mechanical property is stable, and the anti-aging property is excellent.
Owner:NANJING UNIV OF SCI & TECH +1

Artificial intelligence-based archive abnormal behavior analysis and early warning system

This invention relates to the field of digital archives management technology, specifically to an artificial intelligence-based system for analyzing and warning of abnormal archive behavior. The invention uses a graph construction module to transform archive access logs into a relational topology graph; an attribute propagation module performs weighted transfer and accumulation of weights based on a graph traversal algorithm to calculate the cumulative relational index value of nodes; a connectivity retrieval module retrieves cross-regional links based on partition isolation rules; and an early warning output module generates early warning signals based on index values ​​and link status. This invention utilizes the flow mechanism of feature values ​​in a graph structure to aggregate minute risky behaviors within discrete long-term windows, accurately identifying complex abnormal patterns where a single operation may not violate regulations but the long-term cumulative risk exceeds the limit. This solves the problem that existing static statistical rules are unable to detect hidden logical violations, effectively reducing false alarms and false negatives.
Owner:SHANDONG WEIZHUO INFORMATION TECH CO LTD

An energy short-term load prediction method and system based on SE-Block improved transformer

This invention relates to the field of energy forecasting technology, and in particular to a method and system for short-term energy load forecasting based on an improved Transformer using SE-Block. The method includes reversible normalization preprocessing of acquired multivariate load sequence data; feature extraction and fusion of the preprocessed data using improved cross-scale interactive patching, including multi-scale feature extraction, cross-scale interactive alignment, residual correction, and dynamic fusion; and feature filtering of the fused features based on a channel attention mechanism, including feature response based on improved SE-Block and nonlinear interaction of context vectors. This invention addresses the non-stationarity of actual load caused by meteorological conditions and user behavior. By automatically eliminating noise interference among multiple variables, the model accurately depicts the fluctuation details of the load curve, demonstrating its robustness in multivariate load forecasting for integrated energy systems.
Owner:SHANDONG UNIV

Methods for predicting the TCM functions of single Chinese medicinal substances and for drug screening

PendingCN122738679AImprove forecast accuracyEffectively Capture Interaction Forces
This invention provides a method for predicting the TCM functions of a single herbal drug and for drug screening. The method includes: treating each atom in the molecule of the herbal drug to be predicted as a node in a graph; constructing feature vectors for each node based on the physical and chemical properties of each atom; constructing a feature vector matrix based on the feature vectors of all atoms; treating the chemical bonds between atoms as edges in the graph; constructing an adjacency matrix based on the bonding types between atoms; constructing a graph structure matrix based on the feature vector matrix and the adjacency matrix; inputting the graph structure matrix into a graph convolutional neural network model to obtain a function score vector, the length of which is equal to the total number of preset TCM functions, and each score in the function score vector representing the probability that the herbal drug to be predicted belongs to a corresponding preset TCM function; comparing each score in the function score vector with a preset threshold, and determining the predicted TCM function value of the herbal drug to be predicted based on the comparison result. This invention improves the accuracy of function prediction.
Owner:HUBEI UNIV

Poly (p-hydroxystyrene) photoresist film-forming resin and preparation method thereof

The invention relates to poly (p-hydroxystyrene) photoresist film-forming resin and a preparation method thereof, relates to the technical field of photoresist materials, and is used for solving at least one of the problems of relatively wide molecular weight distribution, uncontrollable molecular weight and the like of the film-forming resin prepared by the existing method. According to the method disclosed by the invention, free radical polymerization reaction is carried out under photo-initiation through a reversible addition-fragmentation chain transfer polymerization technology, so that the reversible inactivation reaction of free radicals in a polymerization reaction system is controlled, the molecular weight distribution of the synthesized deep ultraviolet photoresist film-forming resin (poly-p-hydroxystyrene photoresist film-forming resin) is narrow, the PDI is less than 1.2, the reaction is instantly controlled, and the film-forming effect is good. The reaction conditions are mild; the energy consumption is low; and the product is easy to separate and purify.
Owner:BEIJING LUMIRAFT TECHNOLOGY CO LTD

Solid electrolyte precursor solution, solid electrolyte, battery and preparation method thereof

The invention discloses a solid electrolyte precursor solution, a solid electrolyte, a battery and a preparation method thereof, the solid electrolyte precursor solution comprises a lithium salt solution, pentaerythritol glycidyl ether and lithium difluoro (oxalato) borate, and the lithium salt solution comprises a lithium salt and a solvent. The solid electrolyte prepared from the solid electrolyte precursor solution disclosed by the invention has relatively high ionic conductivity.
Owner:SHENZHEN INX ENERGY TECHNOLOGY CO LTD

Water supply sewage treatment filtering device

ActiveCN224194311UImprove rotational driving forceImprove rotation smoothnessFiltration circuitsFiltrationSewage outfall
The utility model relates to a water supply sewage treatment filtering device which comprises a supporting assembly and a filtering barrel rotationally arranged on the supporting assembly, a barrel is coaxially arranged in the filter barrel, an annular filter cavity is formed between the barrel and the filter barrel, an annular sealing plate is arranged above the cavity, a plurality of layers of concentric circle filter materials are arranged in the cavity in the radial direction, and through holes communicated with the annular filter cavity are formed in a barrel body of the filter barrel and the barrel. 4-12 blades are arranged on the upper portion of the inner wall of the barrel in a circumferential array mode, an end cover is coaxially arranged above the barrel through a bearing, a water inlet is formed in one side of the end cover and connected with a water pipe, the water outlet direction of the water pipe is tangent to the inner wall of the side face of the end cover, and tangential water outlet is matched with the blades to drive the filtering barrel to rotate. The bottom of the filter barrel is conical or arc-shaped, and a drain outlet is formed in the center. The device solves the problems that a traditional filtering device is low in efficiency and needs extra power, hydraulic self-rotation driving and multi-stage annular filtering without external power are achieved, the filtering effect is remarkably improved, and the operation and maintenance cost is reduced.
Owner:QUANZHOU MEASUREMENT INST

Centrifugal clutch type drain cover

PendingCN122669773AEfficient aggregation
This invention discloses a centrifugal clutch-type drainage manhole cover, comprising: a supporting base; a manhole cover body, the manhole cover body including a cover body, the cover body having annular and spaced primary drainage outlets and secondary drainage outlets, and an opening and closing grid plate connected to the bottom of the manhole cover body below the primary drainage outlets; and a rotating valve plate assembly, the rotating valve plate assembly being placed below the secondary drainage outlets and connected to the bottom of the manhole cover body. This invention uses gravity as the primary trigger signal and water turbine rotation speed as the secondary trigger signal, achieving normal micro-drainage and increased drainage during heavy rainfall through a filter chassis linked to the opening and closing grid plate. It also utilizes water flow to drive gears for active unblocking. When rainfall further increases, the water turbine drives the centrifugal clutch mechanism to open the secondary drainage outlets, enabling the manhole cover system to automatically complete a three-stage stepped response from micro-drainage to primary strong drainage and then to secondary strong drainage according to the rainfall intensity, ensuring drainage effectiveness.
Owner:JIZE COUNTY HENGYANG CASTING CO LTD

A forest fire smoke detection method based on semantic segmentation

ActiveCN121438321Bcapture moreWide coverage
The application provides a forest fire smoke detection method based on semantic segmentation, and belongs to the technical field of fire warning, and comprises the following steps: feeding a real forest fire data set into a feature extraction network, inputting the extracted features into an improved DeepLabV3+ network for training, and outputting detection and recognition results. According to the complex and changeable forest fire scene, the non-rigid structure, the opacity and the variable mode of smoke and other characteristics, the application deeply studies a feature fusion method, adopts an improved ASPP structure to extract multi-scale spatial features, simultaneously introduces a bottom feature extraction module and a double-path convolution aggregation module to effectively extract smoke boundary information, and uses a lightweight backbone network to replace the original feature extraction network of DeepLabV3+, so that the detection speed is accelerated by using fewer parameters. The application can solve the problem that the false alarm and missed detection rates of the smoke detection method are high, thereby providing technical support for forest ecological protection.
Owner:CENTRAL SOUTH UNIVERSITY OF FORESTRY AND TECHNOLOGY