1176 results about "Semi automation" patented technology

Systems and methods for integrating a continuous glucose sensor, including a receiver, a medicament delivery device, and optionally a single point glucose monitor are provided. Manual integrations provide for a physical association between the devices wherein a user (for example, patient or doctor) manually selects the amount, type, and / or time of delivery. Semi-automated integration of the devices includes integrations wherein an operable connection between the integrated components aids the user (for example, patient or doctor) in selecting, inputting, calculating, or validating the amount, type, or time of medicament delivery of glucose values, for example, by transmitting data to another component and thereby reducing the amount of user input required. Automated integration between the devices includes integrations wherein an operable connection between the integrated components provides for full control of the system without required user interaction.

The invention provides a defect image incremental learning method and system under unmanned aerial vehicle cloud-side cooperation, and the method comprises the steps: constructing a distributed cloud-side cooperation power transmission line defect recognition platform, arranging an inference and semi-automatic labeling system at an edge end, and constructing a multi-model incremental training system at a cloud end through a digital twin technology; the center cloud and the edge nodes cooperate with each other, and the labeling efficiency and the model accuracy can be improved through incremental learning. Meanwhile, samples are marked manually or with the assistance of other systems, and then the model is trained in an incremental mode again. If the incremental training model is significantly improved in accuracy compared with an original model, the incremental training model can be deployed to an edge node to update the original model. According to the method, a mechanism is achieved for the assembly line task in a crowdsourcing mode, the semi-automatic labeling task is divided into a plurality of small assembly line tasks, the labeling work is executed more efficiently through assembly line work, and the obtained labeling result is more accurate by integrating the characteristics of different models.

Fully automated and semi-automated AI autofill grant application processes and associated artificial intelligence (AI) autofill grant application response generation and user supplied information (USI) extraction processes are disclosed for filling in fields of online grant applications with information from manually generated, pre-set text, other sources, and AI-generated information. As a system, the processes relate to an interactive context tool. The system also includes a server, databases, and a large language model (LLM) of an AI system that is trained on the USI from prior grant applications, along with natural language processing (NPL) to understand requested information in fields of a grant application and to generate answers—custom to a particular user or entity seeking a grant—to automatically fill in the fields with the answers corresponding to the requested information.

The invention discloses an RPA-based full-automatic all-electricity invoice issuing method and system. The method comprises the following steps: constructing an RPA system: creating an RPA user, creating a business process, and defining a process rule; data acquisition and preprocessing: capturing basic transaction data, and performing data preprocessing on the captured data; invoicing and intelligent inspection: logging in a full-electricity invoice service platform, filling and submitting data information in an invoicing interface, and verifying various contents of an invoice; invoice circulation and signing determination: sending the issued full-electricity invoice to the user, and monitoring the invoice signing condition; and storage and filing management: uploading the issued all-electricity invoice to a management system. According to the invention, by introducing the RPA technology, the full-automatic processing of the full-electricity invoices from data acquisition and issuing to circulation and archiving is realized, the time consumption of manual operation is greatly reduced, and the invoice issuing efficiency is improved compared with the traditional manual or semi-automatic process.

According to the semi-automatic pushing and connecting equipment and method based on multiple stations, the materials in the material storage device are continuously conveyed to the cut-off assembly through the conveying assembly, then the situation that the materials enter the cut-off assembly is detected through the visual module, and therefore the materials can be reliably conveyed to the pre-inserting device conveniently when the front faces and the back faces of the materials are in different states; then, one side of the FPC connecting end is moved to the bearing assembly through the material supporting device and is adsorbed and fixed through the adsorption device arranged on the bearing assembly, the position of the FPC connecting end is adjusted in the plane through the turnover device to be aligned with the other side of the FPC connecting end, the FPC connecting end is turned over after alignment, pre-insertion is completed through the pre-insertion device, and push-insertion is completed through the push-insertion device; and finally, the pushed and inserted materials and the FPC connecting ends are pressed through the pressing buckle assembly, so that the FPC connecting ends of different sizes can be pushed and inserted, and different materials can be accurately connected with the FPC connecting ends of different models.

The invention provides a building self-correction design method and system based on a big language model, a design experience library and a design rule library are formed by using a BIM model of a historical design scheme and design specifications, and when a new project is designed, the big language model can be used for searching and understanding a historical experience knowledge library, and reasoning ability, so that the design efficiency is improved. Intelligent optimization suggestion is carried out on existing design parameters, the compliance of the suggestion is ensured through retrieval and understanding of a rule base, and finally semi-automatic or automatic correction of building design is achieved.

In one aspect, computer program code is synthesised based on program design artefacts. At least one code synthesis prompt is created based on a program design artefact, and synthesised program code is obtained by submitting the at least one code synthesis prompt to a machine learning (ML)-based generator. The program design artefact for synthesizing the first program code is obtained by submitting at least one program design prompt to the ML- based generator. When the synthesised program code is modified, at least one design update prompt is submitted to the ML-based generator based on the modified program code, causing the ML-based generator to generate an updated program design artefact. Alternatively, when the program design artefact is modified, at least one code update prompt is submitted to the ML-based generator based on the modified artefact, causing the ML-based generator to generate updated program code.

The invention discloses a concrete phase segmentation method based on lightweight deep learning and an SE attention mechanism. The method comprises the following steps: acquiring an original concrete image by adopting X-ray tomography; in the pretreatment process, a semi-automatic labeling method is adopted for accurately labeling the phase of the concrete; dividing the data set into a training set, a verification set and a test set; on the basis of a U-Net network architecture, network lightweight is realized by adopting a strategy of reducing the number of channels of a convolutional layer, and an SE attention mechanism is embedded to construct an SE U-Net model; and finally performing model evaluation on the test set through model training and optimization. According to the SE U-Net model constructed through the method, in a concrete phase segmentation task, mIoU reaches 88.5% and is remarkably improved by 9.7% compared with a U-Net model, meanwhile, the model parameter quantity and computing resource consumption are reduced through lightweight design, and efficient and accurate concrete phase segmentation can be achieved.

The invention discloses an automobile relay assembling device, which comprises an iron core assembling module and a movable spring assembly assembling module, the iron core assembling module comprises a second circulating conveying mechanism, and a seat body feeding mechanism, a movable spring assembly assembling mechanism and a movable spring assembly assembling mechanism are sequentially arranged along the second circulating conveying mechanism; a first circulating conveying mechanism; an iron core feeding mechanism; an iron core hinging mechanism; an iron core hinging mechanism; a first discharging and conveying mechanism; a relay body feeding mechanism is arranged on the side, away from the second circulating conveying mechanism, of the first circulating conveying mechanism. A relay body carrying mechanism is arranged above the position between the second circulating conveying mechanism and the first circulating conveying mechanism. The movable spring assembly assembling module comprises a rotating disc conveying module. A first feeding conveying mechanism and a second feeding conveying mechanism are sequentially arranged along the rotating disc conveying module. A movable contact spring feeding mechanism; a movable spring pin feeding mechanism; a riveting mechanism; and a second discharging transportation mechanism. The whole assembly process is operated by a mechanical device, the assembly refinement degree is high, compared with a manual or semi-automatic mode, the mechanical assembly efficiency is greatly improved, the manual labor force is saved, the assembly precision is improved, and the automation degree is high.

The invention relates to a full-automatic forming system for imitation handmade egg tart skins. The egg tart production line solves the technical problems that most of existing egg tarts are produced in a semi-automatic or pure manual mode, and normal production efficiency is affected. Comprising a tower cup moving plate with a plurality of tower cup bottom die holders, horizontal conveying assemblies are arranged at the two ends of the tower cup moving plate respectively, a tower skin pre-forming part is arranged at one end of each horizontal conveying assembly, a tower skin forming part is arranged in the middle of each horizontal conveying assembly, and a tower skin jacking conveying part is arranged at the end, away from the tower skin pre-forming part, of each horizontal conveying assembly. The tower cup moving plate sequentially penetrates through the tower skin pre-forming part and the tower skin forming part through the horizontal conveying assembly to enter the tower skin jacking conveying part. The device has the advantages that the tower skin preforming part, the tower skin forming part and the tower skin jacking and conveying part are connected through the horizontal conveying assembly, so that the production efficiency of the egg tart skin is improved, the requirement of large-scale production is met, the stable quality of the egg tart skin can be ensured, the manual manufacturing effect is simulated as much as possible, and the defective rate is reduced.

The invention discloses a ground element labeling method and device, equipment and a storage medium, and the method comprises the steps: determining time sequence data corresponding to a target scene, and constructing three-dimensional point cloud data of the target scene based on the time sequence data; based on the three-dimensional point cloud data, determining a bird's-eye view and an elevation map corresponding to the three-dimensional point cloud data, and displaying the bird's-eye view on a user interface; in response to the detected labeling operation for the ground elements in the aerial view, determining first labeling information of the ground elements; and determining a labeling result of the ground element based on the first labeling information and the elevation map. According to the technical scheme, the three-dimensional point cloud data of the target scene is constructed, and the aerial view and the elevation map are determined based on the three-dimensional point cloud data, so that semi-automatic manual labeling is performed at the BEV perspective, and meanwhile, the 3D labeling result of the ground element is obtained in combination with the elevation map, so that the high-quality ground element labeling true value can be conveniently and quickly obtained.

The walnut processing and conveying equipment with the steering mechanism comprises a conveying table, a conveying steering belt, a containing box and a processing assembly, the conveying steering belt is movably installed on the inner side of the conveying table, the containing box is placed above the conveying steering belt, the processing assembly is fixedly installed above one side of the conveying table, and the processing assembly is fixedly installed above the other side of the conveying table. A conveying steering belt is arranged to drive an upper placing box to move, and different processing treatments are carried out when the conveying steering belt moves to the position below the processing assembly; according to the walnut processing equipment, automatic or semi-automatic production can be achieved, the processing efficiency is remarkably improved, the labor cost is reduced, multi-stage processing treatment is conducted on walnuts in the transportation process, parameters such as the temperature and time in the processing process are accurately controlled through arrangement of the temperature controller, and the walnut processing equipment can guarantee the quality and consistency of products; the efficient processing equipment can reduce damage and loss of walnuts in the processing process, and the utilization rate of raw materials is increased.

The invention belongs to the technical field of disc-shaped flange part detection equipment, and discloses rotor inner cavity bolt connection rigidity detection equipment. The rotor inner cavity bolt connection rigidity automatic detection equipment comprises a clamping module, a lifting module, a rotation module and a probe test module which are all integrated on a main body rack, the clamping module is used for clamping and positioning, the lifting module is used for providing pressing force for a test probe and a tested piece, and the rotation module is used for rotating the test probe and the tested piece. The rotation module is used for circumferential indexing positioning among different measurement positions, and the probe test module is used for testing the connection rigidity of a single bolt. All the functions are realized by adopting a semi-automatic means, so that the test working efficiency and the test precision can be improved. According to the automatic detection equipment, a solution of a method for placing a test probe in a space circumferential rotation mode, wherein the test probe is carried in a restricted space on one side, and automatic operation of the evaluation parameter testing and obtaining process of bolt connection in the restricted space of an inner cavity of a disc-shaped part is achieved.

In general, various embodiments of the present disclosure provide systems, methods, apparatuses, technologies, and / or the like for loading, storing, and / or dispensing objects in an automated and / or semi-automated fashion. For example, various embodiments provide a storage system comprising: a frame comprising a shelf; a plurality of doors attached to the frame along a dispensing side, wherein each door can be opened and closed independently through operation of a corresponding release mechanism; a plurality of dividers adjustable to different positions along a length of the shelf to thereby define storage compartments that vary in size along the length of the shelf for storing objects of different sizes; and a controller that, for each storage compartment, is configured to cause operation of the corresponding release mechanism for a series of the plurality of doors corresponding to a dimension of the storage compartment in unison to release an object stored in the storage compartment.

The utility model discloses an assembling structure of a concrete resiliometer, which belongs to the technical field of concrete strength testing and comprises a base, universal wheels are fixedly assembled at four corners of the bottom of the base, a lifting mechanism is fixedly assembled at the top of the base, and an assembling frame is fixedly assembled at the top of the lifting mechanism. Through a semi-automatic testing method, the operation process of the concrete resiliometer can be standardized and simplified, and only the base needs to be controlled to move in a short distance in the testing process; and then the lifting mechanism is controlled to drive the assembling frame to adjust the height position, so that the rebound instrument can be aligned to the corresponding testing point position in the testing area, the overall strength of manual operation is reduced, and the situation that the reading is inaccurate due to the fact that manual fatigue operation is conducted on too many testing point positions is avoided.

The invention discloses a rapid drilling clamp for a carbon fiber badminton racket. The rapid drilling clamp comprises a machining table, a drilling machine, a sliding assembly, a fixing clamp and the like. The sliding assembly horizontally corresponds to the working end of the drilling machine, and the fixing clamp can fix the racket frame. The inner gear in the machining table containing groove is driven by the moving piece to rotate, then the cam is made to rotate through the transmission gear, and the cam abuts against the sliding assembly to be in linkage so as to adjust the position of the racket frame. The scale value of the machining table is matched with the moving part pointer to achieve accurate angle adjustment. The sliding assembly comprises a fixing rod, a first spring, a sliding frame and the like and can horizontally move under the action of the cam. The fixing clamp drives the four fixing assemblies through the driving assembly, and automatic clamping and dismounting of the racket frame are achieved through the second spring and the supporting block. According to the clamp, accurate positioning and rapid drilling of the racket frame are achieved through a mechanical structure, the problems that in semi-automatic machining, manual distance adjusting is tedious, and the intelligent modification cost is high are solved, the clamp is suitable for racket frames of different shapes, and the drilling efficiency and precision are improved.

The invention discloses automatic cleaning equipment for a water inlet pipe of an automobile motor, relates to the technical field of cleaning equipment, and aims to solve the technical problem that when the water inlet pipe of the motor is cleaned, the inner wall and the outer wall are cleaned separately, and the cleaning efficiency is low. A cleaning cylinder is movably connected between the two end plates, and a plurality of draining grooves are formed in the side surface of the cleaning cylinder. By arranging the cleaning mechanism, the interior and the exterior of the motor water inlet pipe can be cleaned at the same time through high-pressure water flow, impurities on the inner wall of the motor water inlet pipe can be scraped off, the clamping and cleaning functions of the motor water inlet pipes of different sizes can be achieved through the clamping assembly, the bent water inlet pipe can be corrected through the straightening assembly, and the interior of the water inlet pipe can be conveniently cleaned; according to the semi-automatic automobile motor water inlet pipe inner wall cleaning device, the semi-automatic automobile motor water inlet pipe inner and outer cleaning function is achieved, and the cleaning effect of the motor water inlet pipe can be improved.

The invention provides a method and a system for constructing a time sequence knowledge graph in the water conservancy field based on a T-ATLOP model, which are applied to the water conservancy field, mainly comprise two key levels, namely a data layer and a knowledge extraction layer, and realize semi-automatic time sequence knowledge graph construction. The method comprises the following steps: carrying out batch document processing on water conservancy census data by utilizing a large language model through a prompt project, crawling water conservancy consultation of water conservancy halls of all provinces and cities by a web crawler, constructing a time sequence knowledge graph data set HE-DOCU through two-step method labeling of large language model assistance and heuristic strategy labeling, and carrying out a knowledge extraction layer part on the basis of an ATLOP model. A new model T-ATLOP is provided for knowledge extraction. And finally, storing the knowledge tetrad of the knowledge extraction layer by using a graph database to form a water conservancy field time sequence knowledge graph. According to the method, the more perfect water conservancy field time sequence knowledge graph is constructed by fully utilizing the water conservancy information, so that the accuracy and the intelligent level of querying related knowledge by hydrological employees can be improved.

The invention relates to the field of computer vision, in particular to an intelligent identifying and sorting method and system for traditional Chinese medicinal materials. Based on a multispectral and microscopic vision fusion engine, millisecond-level authenticity identification of traditional Chinese medicinal materials is realized through sub-pixel-level feature extraction and cross-modal knowledge distillation, and similar medicinal material confusion is thoroughly eliminated by combining PointRend point cloud segmentation and pharmacopoeia map confrontation verification; a rotary YOLO-MoE multi-angle defect detection mechanism is innovated, a zero-adulteration quality control closed loop with zero manual intervention is achieved under the assistance of a self-adaptive guide execution unit, and a map database is continuously upgraded synchronously through a dynamic federated learning mechanism. The technical bottleneck that an existing traditional Chinese medicine identification system depends on artificial experience and is weak in authenticity identification capability is broken through, and the key problems of confusion of medicinal materials with similar forms caused by single dimension of semi-automatic pharmacy visual identification, missing detection of defective products caused by no quality control mechanism and the like are thoroughly solved; and the medicinal material identifying and sorting process is improved to a stable, reliable, self-adaptive and cooperative intelligent level.

The invention discloses a multi-modal fusion AI visual detection method. The method comprises the following modules: a multi-modal data acquisition module, a data preprocessing and aligning module, a multi-modal feature fusion AI module, a real-time decision and feedback module, and a model iteration and data management module. The multi-modal data acquisition module comprises a high-resolution visible light camera, an infrared thermal imager, a high-precision line laser scanner and a synchronous trigger controller; the method further comprises the following steps: step 1, an early-stage preparation stage; 2, a detection execution stage; 3, a decision-making and feedback stage; 4, performing daily maintenance and iteration; the invention aims to provide a multi-modal fusion AI visual detection method, so as to solve the problems that the existing method depends on a single detection mode, only surface texture defects can be identified, and internal problems or three-dimensional form defects of materials cannot be considered; and a traditional detection mode often depends on manual judgment or a semi-automatic process, is low in efficiency and is greatly influenced by subjective factors.

Detecting, locating, identifying, engaging, and / or shifting objects in automated or semi-automated fashion as well as methods, systems, apparatuses, and computer-program products therefor. Embodiments used for shifting objects in automated or semi-automated fashion may be implemented in stationary environments, and / or in moving, e.g., in-transit, environments, and may be used to transfer, route, and / or organize objects based on their designated destinations. The embodiments may further be implemented in a logistics network, thereby increasing the efficiency and capacity of the logistics network, among other benefits.

The present invention provides a forming device and a processing method for a metal T-type connector. The forming device includes a fixed frame, and a slanted top core-pulling assembly is provided on the fixed frame. The slanted top core-pulling assembly includes a support frame, and a first transverse moving assembly and a second transverse moving assembly are symmetrically provided on both sides above the support frame; a third transverse moving assembly and a fourth transverse moving assembly are symmetrically provided on the other two sides above the support frame, and one side of the third transverse moving assembly is connected to a core-pulling part, and one side of the fourth transverse moving assembly is connected to a core-pulling part. The present invention adopts a reusable metal mold core and provides a set of production equipment used in conjunction with the mold core. The production equipment of the present invention greatly improves the efficiency of collecting products after molding by means of slanted top core-pulling, and realizes full separation of the mold core from the product, realizing semi-automatic and efficient production of T-type connectors, effectively improving the batch production process of products, reducing the production cost of enterprises, and improving the economic benefits of enterprises.

The labeling and assembling equipment comprises a conveying line, a material cover supply device, a label supply device, a material cover bearing and positioning device and a turnover device, an assembling station is arranged on the conveying line, and the assembling station comprises a blocking and releasing mechanism and a code scanner; the material cover supply device comprises a material cover discharge end; the label supply device comprises a label generation part and a label supply table; the material cover bearing and positioning device comprises a material cover bearing and positioning table and a correcting and positioning mechanism. The turnover device comprises a picking mechanism with a negative pressure picking part, and the picking mechanism has turnover displacement. According to the invention, the working mode of labeling and transferring the gland after positioning the chip material cover is adopted, so that the condition of assembly deformation is avoided, the labeling position is accurate, and the assembly alignment is reliable. Full-automatic operation can be achieved, and operation is efficient and smooth. Manual semi-automatic operation is achieved through the detection device, so that the utilization rate of labeling and assembling equipment is high, and operation is more flexible. And the rechecking requirement in the full-automatic operation process is met, and operation is stable and reliable.

The invention discloses a causal effect estimation method and system based on a high-throughput medical science data variable, and the method employs the high-throughput medical science data variable to approach to a pseudo-randomization design, so as to obtain unbiased causal effect estimation. Through optimal combination of evidence-based medical design and causal inference theory methods such as new user queue design, L1 regularization tendency score matching, negative control and synthetic positive control, P value calibration and confidence interval calibration, personalized deep causal learning algorithm and the like, a brand new intelligent big data real world research design and analysis process is invented; and an automatic or semi-automatic, universal and robust research design and analysis general solution is provided for realizing evidence-based evidence of high-level observation research by using big data.

The utility model belongs to the technical field of transformer detection, and particularly relates to a withstand voltage test board for a power-taking transformer. Comprising a test board body, a pressing plate connected to the upper part of the test board body through a pressing cylinder, a probe arranged at the bottom of the pressing plate, a test board arranged at the lower part of the test board body, and a plurality of positioning grooves uniformly formed in the top of the test board, the automatic feeding assembly comprises a plurality of positioning blocks evenly arranged on the front side of the testing plate, two clamping plates arranged on the two sides of the positioning blocks and connected through first electric telescopic rods, a transmission lead screw arranged between the two fixing plates, a motor arranged in one fixing plate and a sliding block in transmission threaded connection to the outer wall of the transmission lead screw. According to the utility model, the semi-automatic operation of the withstand voltage test of the transformer is realized, the efficiency of the transformer test can be improved, the risk of electric shock of testers can be avoided, and the detected transformer can be transferred from the test board to the collection table after the test of the transformer is finished.

The invention discloses an electric heating tube machining surface grinding device which comprises a magnetic polishing machine body and a cleaning box, the cleaning box is installed in a machine box of the magnetic polishing machine body, the electric heating tube machining surface grinding device further comprises a maintaining tool and a vibration lifting mechanism, the maintaining tool is installed in a box body of the cleaning box, and the vibration lifting mechanism is installed on one side of the maintaining tool. The maintaining tool is arranged at the cleaning box of the magnetic polishing machine body, the vibration lifting mechanism is arranged in cooperation with the maintaining tool, the maintaining tool can be provided with multiple sets of frame body parts for clamping the electric heating pipes in batches, and then the maintaining tool is fed into the cleaning box at the vibration lifting mechanism in batches for semi-automatic cleaning; the grinding and polishing convenience of the electric heating tubes at the magnetic polishing machine body is greatly improved, the tool is kept not to move when carrying the electric heating tubes to clean in the cleaning box, and the situation that threaded columns of the electric heating tubes are damaged due to mutual collision during grinding and polishing of the multiple sets of electric heating tubes is effectively avoided.

The invention belongs to the technical field of aircraft manufacturing, and discloses an automatic offset compensation boring method for a curved surface normal hole, which directly finishes boring, on-machine measurement and compensation processing of the curved surface normal hole in a machine tool in a semi-automatic manner by using a convenient and universal measuring head without designing a special tool. The technical problems that a part with a curved surface normal hole is long in machining period and large in machining difficulty are solved. Compared with a traditional method, efficient numerical control boring of the curved surface normal hole is achieved, and the machining precision and production efficiency of parts are improved.

The invention provides a construction method and device for a local knowledge graph, and the construction method comprises the steps: extracting a first concept from the multi-text data of a local database, and carrying out the classification and layering of the first concept; and confirming and supplementing the classified and layered first concepts, and optimizing the first concepts into a plurality of second concepts. And segmenting multi-text data into a plurality of text paragraphs in advance, and matching and associating the text paragraphs with the second concept. And respectively extracting a plurality of local concepts and relationships thereof from the associated text paragraphs, and constructing an entity relationship multi-tuple. And integrating similar local concepts in the entity relationship multi-tuple into the same local concept. And analyzing the similarity of all the relationships in the plurality of entity relationship multi-tuples, and evaluating the relevance among all the entity relationship multi-tuples. And based on this, obtaining the constructed knowledge graph. According to the semi-automatic construction method and device, on the basis that expert knowledge is well utilized, the method does not excessively depend on manual construction rules, and the reasoning and understanding ability of a large model is utilized to the maximum extent.

The invention discloses wire harness winding and binding equipment which comprises a rack, a rotating module, a wire drawing module, a binding belt module and a clamping jaw module, wherein the rotating module, the wire drawing module, the binding belt module and the clamping jaw module are arranged on the rack. The rotating module comprises a connector clamping mechanism, a wire harness tensioning mechanism and a rotating mechanism, and can fix a wire harness end connector, support a wire harness to form a ring shape and drive the wire harness to rotate to achieve winding. The wire drawing module guides the wire harness to be uniformly wound on the wire harness tensioning mechanism; the clamping jaw module can take out the annular wire harness from the rotating module and send the annular wire harness to the binding belt module to complete binding belt binding. According to the equipment, through cooperation of all the modules, full-process automation of wire harness winding, uniform wire arrangement, transferring and binding is achieved, the problems that manual or semi-automatic equipment is low in efficiency, poor in winding consistency, unstable in binding and the like are solved, the production efficiency and the product quality are improved, and it is ensured that winding is uniform, the ends are fixed reliably, and binding is firm.

The invention discloses a synthetic defect data generation method based on physical feature fusion, and relates to the technical field of defect data generation, and the method comprises the steps: building a real crack feature parameter library of stamping part cracking defects; a parameterized fractal function is adopted to generate a fracture basic form; the Poisson fusion algorithm is used for achieving material sensing fusion of the crack and the stamping part substrate, and edge artifacts are eliminated; performing personalized adjustment on the initial synthetic defect image to generate final synthetic defect data; and using the generated synthetic defect data for deep learning model training. According to the method, a basic system of a real crack characteristic parameter library and CAE stress field data is constructed, a physical real crack form adaptive to a stress field is generated in combination with a parameterized fractal function, edge artifacts are eliminated through Poisson fusion, personalized adjustment is performed through a semi-automatic tool, defect parameters are fed back and optimized by relying on a deep learning model, and thus the crack quality is improved. And finally generating stamping part cracking synthesis defect data with physical authenticity and deep learning adaptability.