12results about How to "Flexible scalability" patented technology

This application discloses a method and apparatus for constructing a multi-scale adaptive gating MambaPlus network, belonging to the field of artificial intelligence and machine learning technology. The method includes: initializing the network configuration and constructing the basic structure; preprocessing the input data to generate a standard dataset; mapping the input data to the hidden space via an input mapping layer, and extracting backbone features from the Mamba backbone; constructing at least two parallel scale branches in the hidden space to obtain a multi-scale feature set; inputting the backbone features and multi-scale features into an adaptive gating module, dynamically allocating weights and adaptively fusing them through a hierarchical gating mechanism to generate fused features; further enhancing the features through cross-scale attention and feedforward enhancement, and then superimposing the residuals to generate the final discriminative features; finally, completing category prediction and model training evaluation. This application, while retaining the advantages of Mamba's long-range dependency modeling, addresses the problems of insufficient utilization of multi-scale information, poor adaptive feature fusion, and low robustness in complex scenarios.

This utility model relates to the field of cold storage structure technology and discloses a novel prefabricated cold storage unit, including a cold storage frame composed of a first frame, a second frame, and a connecting frame. Each of the first frame and the connecting frame has a door on opposite sides, and the connecting frame is positioned between the first and second frames. A partition is slidably connected inside the frame to change the internal space of the cold storage. A fixing component is positioned between the frames to facilitate assembly; the fixing component includes a plug-in plate and grooves formed inside the first and connecting frames. This utility model adopts a modular structure design, allowing for flexible expansion and combination of cold storage capacity through the splicing of the first frame, second frame, and connecting frame, meeting diverse application needs. The structure of slides and fixing seats ensures the stability of the longitudinal assembly. The use of plug-in plates, grooves, and fixing bolts makes overall assembly quick, disassembly convenient, and transportation and maintenance easy.

The application provides a linear motion test system of a multi-channel inductive proximity sensor, relates to the technical field of aviation sensor testing, and comprises a base, a plurality of linear motion platforms installed on the base, a plurality of targets fixed on the linear motion platforms through target supports, a plurality of proximity sensors fixed on the base through sensor supports, a plurality of stepping motors for driving the linear motion platforms to move along a straight line, and an upper computer connected with the stepping motors through a CAN bus via a CAN distributor, for sending position control instructions to the stepping motors, so that the stepping motors drive the linear motion platforms to drive the targets to move, the distance between the targets and the proximity sensors is adjusted, the inductance values of the plurality of proximity sensors are collected by a measured controller, and performance verification is completed. The application realizes automatic and accurate control, reproduces real working conditions, improves the reliability of test data, adapts to the testing requirements of multiple scenes, and guarantees the operation reliability of aviation key systems.

The utility model discloses an earphone with a detachable NFC (Near Field Communication) magnetic patch, which comprises an earphone main body and an NFC magnetic patch, and the NFC magnetic patch is detachably arranged on the earphone main body; an NFC chip is arranged in the NFC magnetic patch and is used for storing and transmitting NFC data; the earphone main body is internally provided with a PCBA board with an NFC processing module, and the PCBA board is used for processing NFC signals. According to the utility model, through the design of the detachable NFC magnetic patch, convenient interaction between the earphone with the NFC magnetic patch and a mobile phone APP is realized, and the intelligent degree of the earphone and the user experience are improved; a user can replace or upgrade the NFC magnetic patch according to needs, and flexible expansion and upgrade of earphone functions are realized.

This invention discloses a method, system, and storage medium for multivariate time-series feature extraction and grade prediction in flotation processes. The method includes the following steps: Step S1: Raw data input and encoding; encoding and structured input of time-series data composed of various process variables; Step S2: Extracting dynamic features; Step S3: Condition-guided encoding modeling; Step S4: Enhancing the saliency of key variables and important time segments; employing a multi-head attention mechanism to match key-value pairs generated from target-guided query vectors and multi-scale features; Step S5: Outputting a prediction module; performing feature fusion and nonlinear mapping on the attention mechanism output to output the predicted concentrate grade and recovery rate for future times. The system and storage medium are both based on the above method. This invention has advantages such as higher intelligence, better controllability, and improved prediction accuracy and model adaptability for key indicators in the flotation process.

The utility model discloses a water tank empty real exchange goods shelf device relates to the technical field of logistics operation, wherein, water tank empty real exchange goods shelf device includes frame, conveying mechanism, first push mechanism, second push mechanism and elevating system, and conveying mechanism includes first conveying assembly and second conveying assembly that first conveying assembly and second conveying assembly are spaced apart on the frame along the vertical direction, and elevating system is used for moving material from the discharge end of second conveying part to the feed end of third conveying part. The utility model discloses the technical scheme through adopting four oblique settings conveying part, forms the continuous goods box circulation path, and effectively simplifies the conveying flow, and improves work efficiency, through adopting first push mechanism and second push mechanism, can realize the accurate transfer of goods box between different levels, through adopting elevating system, realizes the automation exchange of empty box and real box, ensures that empty box can be in time back to logistics system and continue using, guarantees the stability and continuity of production rhythm.

The invention discloses an urban traffic flow prediction and signal optimization method based on a space-time diagram neural network. The method specifically comprises the following steps: S1, constructing an urban road network topological graph; s2, multi-source heterogeneous data fusion and feature extraction; s3, establishing a space-time diagram neural network prediction model; s4, a signal optimization algorithm based on a prediction result; s5, model training and online strategy updating; s6, integration and real-time reasoning optimization are carried out; according to the method, through constructing topological graph representation of an urban road network, fusing multi-source heterogeneous traffic data and adopting a deep learning architecture combining a graph convolutional network and a time sequence prediction model, accurate short-term prediction of traffic flows of intersections and road sections is realized, and a dynamic signal optimization algorithm is designed based on a prediction result, so that the traffic flow prediction efficiency is improved. The adaptive adjustment of traffic signal timing is realized, so that the traffic efficiency of a road network is improved, and the vehicle delay is reduced.

A back-emitting lidar transmitter chip based on a nanobeam switch array includes a nanobeam switch routing network, a metal reflective layer, and a metasurface lens serving as a Fourier transform lens. The nanobeam switch routing network is fabricated on the front side of the chip. After the input optical signal enters the nanobeam switch branch of the network, it is routed by specific nanobeam switch units to the corresponding nanobeam emission branch, and then emitted into free space by specific nanobeam emission units. The front metal reflective layer reflects the upward-radiated light waves downwards, and finally, all downward-emitted light waves are collimated and emitted in a designated direction by the back metasurface lens. This invention enables directional transmission, simultaneously achieving switch selection and vertical emission functions. The silicon-based metasurface lens can serve as a Fourier transform lens to achieve high-precision, wide-range beam scanning, providing a superior technical path for the large-scale manufacturing of silicon-based integrated lidar transmitter chips.

This application provides a method and system for unstructured text parsing and question answering based on a large language model. The method includes: acquiring raw unstructured text data from a data source at preset intervals; performing deep semantic analysis on the raw unstructured text data using a large language model to extract key entities and data structures; dynamically generating or updating an existing JSON structure template based on the key entities and data structures; converting the raw unstructured text data into standard structured data according to the JSON structure template; storing the structured data in a time-series database and establishing a multi-dimensional index; obtaining time information and query intent from natural language query information input by the user; and generating question-answering results or trend analysis results using a large language model based on the query intent and the structured data obtained from the time-series database according to the time information.

This utility model discloses a modular integrated control cabinet, relating to the field of control cabinet technology. The modular integrated control cabinet includes a cabinet body and a cabinet door connected to the cabinet body, and further includes a power module assembly, a control module assembly, a signal module assembly, a module expansion assembly, an isolation assembly, and a cable tray assembly. The cabinet body includes a support frame, the outer layer of which is fixed with galvanized steel plates and aluminum alloy steel plates by bolts. The modular design of this utility model makes the internal structure of the control cabinet clear, facilitates the installation and maintenance of electrical components, and allows for flexible configuration and expansion of functional modules according to actual needs, improving the versatility and adaptability of the control cabinet. This utility model, through the cabinet frame with two layers of steel plates and the good cable management of the cable tray assembly, improves the stability, reliability, and electromagnetic compatibility of the control cabinet, and extends the service life of electrical components.

The invention belongs to the technical field of optical fiber heat flux density sensing, and discloses a gradient strain coupling FBG heat flux density sensor preparation method, which mainly comprises an optical fiber microgroove processing technology, a metal bonding layer preparation technology based on a magnetron sputtering technology and a gradient strain coupling sensitive layer preparation technology based on an electroplating technology. The method is characterized in that a microgroove is prepared on the surface of the optical fiber through a hydrofluoric acid wet etching process to improve the bonding strength with a metal layer; a first metal layer with high bonding strength is prepared on the surface of the microgroove optical fiber through a magnetron sputtering process; and a gradient strain coupled second metal layer is prepared on the surface of the magnetron sputtering metal layer through a variable-diameter electroplating process. According to the method, the optical fiber microgroove structure and the oxyphilic metal material are corroded through hydrofluoric acid to improve the bonding strength of the optical fiber and the metal, the FBG spectrum is broadened through the second metal layer subjected to gradient strain coupling, the heat flux density sensing performance is optimized and improved, the method can be suitable for machining of various metalized FBG metal layer structures, the process flexibility is high, and the application range is wide. The depth of the microgroove, the thickness of each metal layer and the gradient strain characteristic of the outer metal layer can be flexibly controlled, the heterogeneous material bonding strength of the special metallized fiber is effectively improved, and the preparation cost of the special metallized FBG is reduced.

This utility model discloses a transfer device for packaging injection-molded toys, relating to the field of injection-molded toy production and packaging technology. The utility model includes a conveying assembly comprising a frame, a conveyor belt disposed inside the frame, a motor for driving the conveyor belt, a side plate fixed to the upper side edge of the frame, a top plate fixed to the top of the side plate, a vision sensor detector installed in the middle of the top plate, and a pull-out collection plate disposed on the lower side of the frame. The collection plate is arranged parallel to the bottom of the conveyor belt, and the detection end of the vision sensor detector is disposed above the conveyor belt, and the vision sensor detector is connected to a remote control terminal via a signal line. By adding a pull-out collection plate and using an edge-folding structure, this utility model can not only effectively collect products or waste materials accidentally dropped during the transfer process, preventing material spillage and pollution of the working environment, but also facilitates regular cleaning and maintenance.