20results about How to "Improve correspondence" patented technology

The application relates to the technical field of computer-aided design, and discloses a similar part recommendation method based on CAD context perception, which acquires a target part object selected currently, triggers a similar part recommendation process when the target part object changes, extracts feature information of the target part object, constructs context information and generates semantic description information, generates a geometric feature vector based on the target part object, generates a semantic feature vector based on the semantic description information, fuses the two vectors, and obtains a comprehensive feature vector, calculates the similarity between the target part object and each candidate part in a part library according to the comprehensive feature vector, determines a similar part according to the similarity calculation result, and outputs a recommendation result. The application can improve the corresponding degree of similar part recommendation and a current design context, improve recommendation accuracy and part reuse effect.

The application discloses a centrifugal pump volute flow channel parameter optimization method and system, and relates to the technical field of industrial equipment digital optimization; the method comprises the following steps: based on volute structure data and mechanical seal installation data, the inner wall of the volute is partitioned and identified to obtain a sealing front area, a pressure transmission area and a flow discharge area; based on the sealing front area, the pressure transmission area, the flow discharge area, mechanical seal contact requirement data and target working condition data, contact control analysis is carried out to obtain contact control data; based on the sealing front area, the pressure transmission area, the flow discharge area, the contact control data and the volute structure data, flow channel adjustment item extraction is carried out to obtain regional adjustment data; based on the regional adjustment data, impeller outlet structure data, the contact control data and the target working condition data, flow channel sequence correction is carried out to obtain target flow channel parameters; the application can take into account the sealing working state and the target working condition requirement, and improve the pertinence and adaptability of centrifugal pump volute flow channel parameter optimization.

The application provides a kind of urban water environment water quality synchronous detection device and detection method based on space-time spectrum strong association;Device includes excitation light source module, light splitting modulation module, micro-volume flow-through sample cell, fluorescence receiving module, transmission receiving module, scattering receiving module, physical and chemical parameter synchronous detection module, data acquisition control module and data processing module;The application constructs space-time spectrum joint data set based on synchronous data acquisition, realizes complex mixed response decoupling through space-time spectrum strong association constraint and physicochemical-optical coupling constraint, and realizes quantitative analysis of key biochemical characteristic elements of water body based on pure pixel response spectrum extraction and non-negative quantization.The application can reduce sample consumption, shorten response time, and realize rapid detection, accurate identification and dynamic monitoring of key physicochemical properties and optical behavior characteristics of water body, and has advantages such as easy cleaning and maintenance of sample cell, low risk of cross contamination, high synchronicity and strong interpretability of analysis results.

This invention relates to the field of power material management technology, and discloses an auxiliary management device and method for power spare parts, including a control platform module for maintaining the binding relationship between slots and specifications, maintaining the correspondence between turnover containers and container types, maintaining the candidate specification set for isolation batches, and generating a restricted regression sequence; a specification self-certification slot module for storing spare parts in separate slots and binding them to specifications; a turnover container module for carrying spare parts and locking them to the specification self-certification slot module; an access control module; a handheld terminal module; a mixed placement isolation port module; and a signature chain generation module. By introducing a structural docking and docking locking mechanism between the specification self-certification slots and turnover containers, turnover containers can only dock in specification self-certification slots that match their container type, restricting the storage and return path of spare parts, thereby effectively reducing the risk of specification confusion caused by misplacement or incorrect placement.

This application relates to a cleaning module, a smart toilet seat, and a smart toilet. The smart toilet includes a toilet seat and a smart toilet seat, with the smart toilet seat covering the toilet seat. The smart toilet seat includes a core base and the cleaning module, which is located on the core base. The cleaning module includes a base, a spray washing assembly, and a drying tube. The spray washing assembly is located on the base and is capable of telescopic movement relative to the base. One end of the spray washing assembly has a water spray nozzle. The drying tube has a drying air duct inside, and its two ends have an air inlet and an air outlet, respectively. The drying tube is located on the spray washing assembly and is arranged parallel to it. The end of the drying tube with the air outlet is close to the end of the spray washing assembly with the water spray nozzle. The drying tube is capable of telescopic movement relative to the base under the action of the spray washing assembly.

The application provides an encoder training method, an audio generation method and an audio retrieval method, and relates to the technical field of audio processing. The encoder training method comprises: adjusting an audio quantization encoder to obtain a hierarchical encoder; wherein the hierarchical encoder is provided with an associated semantic codebook and an audio codebook; and jointly training the hierarchical encoder and a text encoder based on a joint loss to obtain an optimized encoder; wherein the hierarchical encoder is used for quantization compression processing of audio data to obtain fusion features, the fusion features comprising semantic features and audio features; the text encoder is used for processing text information associated with the audio data to obtain a text vector; and the fusion features and the text vector have relevance. The audio generation method comprises: determining a generation requirement; wherein the generation requirement comprises a semantic requirement and an audio requirement; and processing based on the generation requirement by using the optimized encoder to obtain requirement audio data.

The utility model discloses a water conservancy pipeline connection auxiliary device which comprises a fixing base, fixing stand columns are fixedly connected to the four corners of the top end of the fixing base, a lifting frame is movably connected to the middles of the fixing stand columns, a first threaded rod is rotatably connected to the middle of the lifting frame, and a sliding plate is movably connected to the interior of the lifting frame. A first threaded rod penetrates through the middle of the sliding plate and is connected with threads, a lifting driving assembly is fixedly connected to the middle of the fixed base, walking wheels are fixedly installed at the four corners of the bottom end of the fixed base, a moving assembly is fixedly connected to the top end of the sliding plate, and a clamping and fixing assembly is installed in the moving assembly. A supporting assembly is fixedly connected to the top end of one side of the lifting frame, a transmission assembly is movably connected between the supporting assembly and the moving assembly, when the device is used, the positions of the clamping frame and the bearing frame can be adjusted, installation and position adjustment of a pipeline are facilitated, the stable fixing effect is guaranteed, and the pipeline connecting quality is guaranteed.

The application discloses a kind of fireproof performance prediction systems of gypsum-based fireproof coating, it is related to gypsum production technical field, including: unit generation module, for establishing continuous development path along the continuous surface relationship between the adjacent positions of target component surface, identify geometric mutation center, connection structure center or coating turning center, and with any one in geometric mutation center, connection structure center or coating turning center as starting position, expand to the adjacent surface area around starting position continuously covered by gypsum-based fireproof coating, to the place where continuous coating boundary terminates, connection structure transition or geometric mutation termination stops, generates local prediction unit.The application establishes continuous development path along the surface of target component, identifies geometric mutation center, connection structure center or coating turning center, and generates local prediction unit accordingly, so that the fireproof performance prediction object is changed from overall area or regular area to local prediction unit constructed around complex key parts.

The invention provides a renal artery stenosis ultrasonic diagnosis report automatic generation method, and relates to the technical field of image data processing, and the method comprises the steps: inputting a plurality of renal artery ultrasonic images, carrying out the redundancy processing, calculating the semantic correlation between each image and a diagnosis text, and screening out a key image; performing feature extraction and enhancement on the screened key images, constructing a static medical knowledge memory matrix, and aligning enhanced visual features with the static medical knowledge memory matrix to generate knowledge alignment vectors; maintaining a dynamic memory network, and injecting the knowledge alignment vector into the dynamic memory network; and fusing the external knowledge token into a code, and finally outputting a structured renal artery ultrasonic diagnosis report based on the enhanced visual features, the external knowledge token and the dynamic memory network state. According to the method, redundant information processing and fine-grained feature extraction of the renal artery ultrasonic image are realized, and external medical knowledge is guided to generate a long ultrasonic diagnosis report which is consistent in logic and has clinical interpretability.

The utility model discloses a portable balcony photovoltaic module support structure belongs to photovoltaic technical field, can carry out assembly in advance in factory, and when packing, folding, when installing, stretch again, try to reduce the workload of user self -assembly. It includes stand, the oblique beam of upper end articulates with stand, the oblique support component of first end articulates with oblique beam and second end articulates with stand, install photovoltaic module on the oblique beam, the stand is U type spare, the upper end of oblique beam articulates in the inside of stand, the oblique support component includes the first side bracing plate and second side bracing plate of setting first transverse parallel, the second end of first side bracing plate and second side bracing plate articulates respectively in the transverse both sides of stand, the oblique support component is separated with oblique beam and is in the storage state respectively after, the oblique beam is stored in the inside of stand under the storage state, the oblique support component under the storage state, first side bracing plate and second side bracing plate are respectively and side by side setting in the transverse both sides of stand.

The high-precision double-color injection molding positioning device comprises a rotating plate, the top end of the rotating plate is fixedly connected with a connecting block, one side of the connecting block is fixedly connected with a connecting rod, the end, away from the connecting block, of the connecting rod is rotationally connected with a rolling wheel, and the interior of the rolling wheel is rotationally connected with a sliding rod; a fixing plate is arranged at the bottom of the rolling wheel, a guide groove is formed in the side, close to the connecting rod, of the fixing plate, and a connecting plate is fixedly connected to the side, close to the rotating plate, of the fixing plate. The rotating plate and the fixed plate are connected through the connecting rod and the roller, the guide groove with the central angle of 180 degrees is formed in one side of the fixed plate, and when the rotating plate drives the first male die and the second male die to rotate, the connecting rod drives the roller to rotate on the surface of the fixed plate, and the sliding rod rotates in the guide groove, so that rotation of the rotating plate is guided; and when the sliding rod rotates to one end of the guide groove, the pressure sensor detects the sliding rod, so that high-precision positioning of the mold is realized.

This utility model discloses a degassing device for geological logging, comprising a degassing tank body, a stirring mechanism, and a centrifugal mechanism. The degassing tank body is cylindrical, including a tank cover and a tank body. The tank cover is located at the top of the degassing tank body, and the tank cover and the tank body are sealed together by a sealing element. The tank body has a vertically arranged partition, which divides the tank body into a separation chamber and a stirring chamber. The bottom of the partition has a hollow support structure. The centrifugal mechanism and the stirring mechanism are respectively arranged corresponding to the separation chamber and the stirring chamber. The top of the tank cover is provided with an inlet pipe and a collection pipe. The inlet pipe is connected to the separation chamber, and the collection pipe is connected to the stirring chamber and close to the top of the stirring chamber. The inlet pipe has a feed control valve. This utility model, through the synergistic action of the centrifugal mechanism and the stirring mechanism, utilizes the dual effects of the shear force of stirring and centrifugal force to enable natural gas to be separated from the drilling fluid more quickly and thoroughly, significantly improving the degassing efficiency.

The invention discloses an image-text recognition method and device based on deep learning, and a medium, and relates to the technical field of image-text recognition, and the method comprises the steps: carrying out the scale normalization of a to-be-recognized image, calculating a gradient magnitude, obtaining a skeleton diagram, and carrying out the projection convergence in a width direction, and generating a time sequence stroke energy sequence; inputting the normalized image into a shared coding network to obtain a time sequence feature sequence, constructing a weight in a height dimension according to the skeleton graph, performing weighted aggregation on the feature graph, and obtaining an aggregation feature aligned with a width time sequence; constructing a CTC branch and an autoregression branch based on the aggregation features, and outputting an emission probability sequence, a character probability and attention distribution; determining an alignment path according to the emission probability sequence and acquiring a character time interval, executing constrained beam search by taking the character time interval and the time sequence stroke energy sequence as constraints, and outputting an identification text sequence by combining CTC sequence scores and autoregression scores. Timing alignment is enhanced and alignment drift is suppressed.

This invention discloses a method and system for visual analysis of big data in enterprise operations, specifically relating to the field of enterprise operation data processing and visualization analysis technology. It includes acquiring target action records, edge fact records, and cloud fact records. Each edge and cloud fact record includes a business object identifier, a fact item identifier, a fact value, an observation time, and a reception time. By extracting the edge-seen facts and cloud-seen facts at the time the target action occurred, backfilled facts formed before the action occurred but received by the cloud after the action occurred are identified. Based on these backfilled facts, the cloud-time records are corrected and visually evaluated.

The invention discloses a plug which comprises a main body, the main body is provided with two electricity taking contact pieces which are arranged at an interval so as to be suitable for being connected with a socket in an inserted mode to take electricity, and a shifting piece is arranged in the main body; the load wire is arranged on the main body and is suitable for being connected with a load electric appliance; the heat conduction structure is arranged in the main body to conduct heat generated by the load wire and the electricity taking contact piece to the main body; the thermal protector is arranged at the joint of the load wire and the electricity taking contact piece, and the thermal protector comprises a bimetallic strip; wherein the load wire and the electricity taking contact piece are connected and conducted through the shifting piece, and when the temperature of the bimetallic strip is larger than or equal to the preset temperature, displacement can be generated, and the shifting piece is driven to be disconnected. Through the heat conduction structure, the overall temperature rise of the plug is balanced, local overheating is avoided, accurate sensing of the overall temperature is realized, the overall structure is simplified, the cost is reduced, reliable protection is realized, the plug is adaptive to a high-power long-time power-on scene, fusing and firing are effectively prevented, and the safety and the service life of a high-power electric appliance or long-time power utilization are improved.

The utility model provides a kind of battery bottom penetration welding machine, including processing table, the surface of processing table is equipped with two groups of electric drive feeding and discharging track, feeding and discharging track is equipped with fixture assembly, the fixture assembly includes mounting bracket, it is easy and fast to operate, and welding quality is reliable, and work reliability is high, and equipment overall structure is simplified stable, and floor space is small, satisfy customer site demand.Adopt double-station alternate manual feeding and discharging, module is conveyed to welding position, battery positioning fixture automatic positioning overturn makes battery bottom upward, galvanometer XY two axes automatic alignment welding, reduce labor intensity, guarantee the consistency of welding quality, battery shell is not damaged after welding, welding surface is smooth and even, no obvious splashing, no false welding, overwelding phenomenon, tightly weld together with shell bottom of roll core, ensure that battery stable operation under various complex conditions, prolong battery service life, guarantee the security of battery use.

This invention discloses a method for optimizing a geo-generative engine and intelligent recommendation retrieval ranking based on a large model, relating to the fields of intelligent geographic information processing and artificial intelligence. This invention cleanses, normalizes, and extracts features from user geo-demands to form standardized demand information; it generates contextualized geographic content based on the GeoGeoLM model combined with multidimensional geographic data and cross-modal features; it recalls candidate points through multi-granularity geographic hierarchical indexing, and achieves comprehensive ranking by combining semantic matching, spatial matching, and dynamic weight allocation, followed by diversity-aware re-ranking; it outputs the generated content in association with the ranking results, and iteratively optimizes based on anomaly information and user behavior data. This method can improve the scenario adaptability of geo-content generation and retrieval ranking.

This invention discloses a water purification membrane and its preparation method, relating to the field of membrane material technology. In preparing the water purification membrane, medical absorbent cotton is used as the cellulose raw material. The cotton undergoes pretreatment and allylation. Pretreated medical absorbent cotton and allyl cellulose are mixed in a suitable ratio and dissolved in lithium chloride solution to form a spinning solution. Ferric chloride is then added for electrospinning. A 0.1 mol / L sodium hydroxide solution is used as the coagulation bath, allowing ferric chloride to diffuse to the outer surface of the fiber during spinning, generating ferric hydroxide while simultaneously forming a rich porous structure within the fiber, resulting in a fiber membrane. The fiber membrane is then crosslinked with polyethylene glycol diacrylate and a photoinitiator, and finally treated with pyrrole hydrochloride solution to obtain the water purification membrane. The water purification membrane prepared by this invention exhibits excellent mechanical properties, water flux, and filtration effect.

The application discloses a caterpillar type harvester path automatic tracking method and system, and relates to the technical field of vehicle path tracking control; boundary object set is generated by obtaining operation boundary data, header boundary data, machine motion data and caterpillar execution data, and boundary chain is formed by filtering according to the direction connection, adjacent connection and header left and right end position change of the boundary object; the header passing area is demarcated along the boundary chain, the advancing zone is limited in combination with the relative relationship of the header, and the caterpillar control amount is generated according to the advancing zone, the machine motion data and the caterpillar execution data; after the execution of the caterpillar control amount, the next caterpillar control deviation is corrected by using the machine, the caterpillar and the header boundary feedback; when the boundary chain is interrupted or the connection is abnormal, the connection boundary is generated, and the boundary chain is connected back after the boundary object is obtained again. The path tracking object is kept corresponding to the actual harvesting boundary of the header, and the continuity of the path tracking control under the boundary abnormal scene is improved.

The application provides a reference transformer-based data super-resolution reconstruction method, device and system, and relates to the technical field of data processing.The reference transformer-based data super-resolution reconstruction method comprises the following steps: obtaining to-be-transformed low-frequency load data, performing logarithmic sum and normalization processing on the to-be-transformed low-frequency load data, and performing data division according to a conversion sequence length to obtain a preprocessed to-be-transformed low-frequency load sequence; inputting the preprocessed to-be-transformed low-frequency load sequence into a qualified data super-resolution reconstruction model to obtain reconstructed super-resolution high-frequency data; through the design of a separate feature extraction module and an information supplement module and the feature fusion mode of low-frequency load features, the corresponding relationship between low-frequency data and high-frequency data is strengthened, and feature expression is more comprehensive; in addition, continuous reconstruction and periodic reconstruction fully utilize the change characteristics of data, fully express long-term features of data, reduce the loss of data details, and improve the accuracy of data reconstruction.