29 results about "Box model" patented technology

The embodiment of the invention provides a cargo boxing sequence generation method and device, electronic equipment and a storage medium, and belongs to the technical field of computers. The method comprises the steps that a to-be-loaded goods list is acquired, and the to-be-loaded goods list comprises a plurality of original goods; obtaining a container body model of the target container and a preset boxing constraint condition; based on the boxing constraint condition, the original goods and the box body model, scheme planning is conducted, and a goods boxing scheme is obtained; the cargo boxing scheme is disassembled, and a cargo boxing guide is obtained; wherein the cargo boxing guide is used for indicating the position and sequence of loading the original cargoes into the target container. The embodiment of the invention can improve the feasibility of a cargo boxing scheme.

The invention discloses a public pipe gallery collision detection method, and belongs to the technical field of pipeline collision detection and three-dimensional simulation. The method comprises the following steps: constructing a spatial geometric model of a public pipe gallery according to gallery entering design, and extracting vertex coordinates and triangular mesh information of the spatial geometric model; a mixed hierarchical bounding box model is established, a shaft alignment bounding box is adopted for static structures such as a pipe shaft, a pipe frame and monitoring equipment, a capsule bounding box is adopted for a pipeline, and a hierarchical bounding box tree is constructed from top to bottom; performing interference judgment between the bounding boxes by adopting a repeated projection method to realize coarse detection; performing triangular mesh point cloudization on intersected leaf nodes, and performing fine detection by calculating an inclusion relation between a point set and a capsule bounding box; and finally outputting a collision position. According to the invention, through combination of the mixed hierarchical bounding box structure and the triangular mesh point cloud technology, the precision and efficiency of collision detection in a public pipe gallery scene are significantly improved, the false alarm rate and the calculation complexity are effectively reduced, and the method is suitable for real-time collision detection and pipeline entry planning of a complex industrial pipe gallery.

The invention discloses a cargo boxing method and device, a storage medium and electronic equipment. Comprising the steps of obtaining point cloud data of a target material box in response to received current cargo boxing task information; based on the point cloud data of the target material box, a virtual material box model corresponding to the target material box is established, and the virtual material box model comprises a goods storage position for placing target goods; a virtual cargo model corresponding to the target cargo is established, the virtual cargo model corresponding to the target cargo is placed at a matched stock position in the virtual workbin model based on workbin historical data, and a virtual workbin model after boxing is obtained; and based on the boxed virtual material box model, a control instruction set is generated, and the control instruction set is used for controlling the transportation equipment to place the target goods at the corresponding preset position in the target material box. The technical problems that in the traditional boxing process, efficiency is low, and accuracy is insufficient are solved.

The application relates to a black box model inversion attack method and device for a text classification model, wherein the method comprises the following steps: extracting a text sequence beginning fragment in a public data set, and constructing an auxiliary data set according to the text sequence beginning fragment; taking the text sequence beginning fragment in the auxiliary data set as a query to access an attack model, generating a complete text sequence, scoring the complete text sequence, obtaining a text sequence score satisfying a preset maximization condition, taking the text sequence score satisfying the preset maximization condition as a target to iteratively optimize parameters of the attack model, and generating a trained attack model; generating a text sequence-score pair set by using the trained attack model; based on the text sequence-score pair set, filtering out a text sequence greater than or equal to a preset score threshold, and generating a final inversion attack result of the attack model according to the text sequence. Therefore, the problems of few sources of model inversion attack information and high attack difficulty of the model inversion attack on the text classification model in a black box scenario are solved.

The invention provides a packaging box bursting strength detection device and a detection method, and mainly relates to the field of material detection. A packaging box bursting strength detection device comprises a bursting strength testing machine, a composite environment test box and a transfer mechanism, a humidity module, a warming module and a temperature and humidity detection sensor are arranged in the composite environment test box, an indexing rotating mechanism is arranged at the bottom of the composite environment test box, and a center shaft is arranged at the top of the indexing rotating mechanism. A plurality of layers of bearing frames are arranged on the center shaft in a sliding mode, a plurality of bearing discs are evenly arranged on each layer of bearing frame in a circumferential mode, the transferring mechanism comprises an action unit and a grabbing unit, and the grabbing unit is used for grabbing samples in the composite environment test box. The method has the beneficial effects that different humidity environments are created, the burst strength data of the carton in the different humidity environments are obtained, reference is provided for packaging box model selection in the special humid environment, and the situation that goods transportation and storage are affected after the humid strength of the packaging box is reduced is avoided.

The invention discloses a structural reliability optimization design method based on a probability and probability box hybrid model, and the method comprises the steps: building a solving model of a reliability index through employing a quadratic four-moment method based on a maximum entropy principle, so as to improve the calculation efficiency of reliability analysis; the classification capability of a K-nearest neighbor algorithm model is adopted to replace a reliability evaluation process in constraint conditions, and design variables are classified so as to reduce the calculation cost of the constraint conditions in the optimization design process; according to the method, a complex nested optimization design problem can be converted into a single-layer deterministic optimization design problem, the optimization calculation efficiency can be improved essentially, and the method has wide engineering application value in the field of engineering mechanical structure design.

The invention provides a production design method for a case of a laser cutting machine, which comprises the following steps: acquiring an initial case model, and processing the initial case model through a topological optimization algorithm to generate an optimized skeleton structure; according to the optimized skeleton structure, finite element analysis is adopted to simulate vibration source influence, inherent frequency distribution is determined, and a frequency matching result is obtained; adjusting the skeleton structure parameters according to the frequency matching result, and planning a functional channel path to obtain channel path data; for the channel path data, extracting force transmission path information, and generating a conflict detection result; modifying the channel path data according to the conflict detection result, and obtaining a conflict-free integrated structure model; simulating deformation data under a multi-working-condition load through the integrated structure model, judging whether the deformation data exceeds a preset threshold value or not, and obtaining a deformation evaluation result; according to the deformation evaluation result, lightweight parameters are adjusted, and a final case design scheme is generated.

The application discloses a substation operating box modular design system and an operating box, and belongs to the field of substation secondary equipment design. The system comprises: a substation unit for determining the number of operating boxes and the functional design requirements and constraint conditions of each operating box; a coding unit for performing bit type spectrum coding on each operating box according to the functional design requirements and constraint conditions; a module selection unit for selecting corresponding operating box models according to the bit type spectrum coding and selecting corresponding functional modules from a preset functional module library; and a module assembly unit for determining the installation positions of the functional modules in the operating boxes. The application can realize the standardized, modular and automated design of the substation operating box, effectively reduce the design and production costs, and facilitate the maintenance and management.

This invention relates to the field of artificial intelligence security and discloses a method for generating adversarial examples for voiceprint recognition. It integrates multiple existing voiceprint recognition models to generate a substitute model to replace the target black-box voiceprint recognition model. Adversarial examples are generated by attacking the substitute model, thereby attacking the target black-box model. This overcomes the difficulty of obtaining model information from a black-box model and improves the low success rate of attacks. In generating adversarial examples, a Nesterov-based accelerated gradient method is used, which can find adversarial examples with better attack effects more quickly. During the generation of adversarial examples, the target black-box model is queried a small number of times to correct the direction of adversarial example generation, thus improving the success rate of the attack.

The present disclosure relates to a system and computer-implemented method for thermal characterization of buildings to assist in thermal assessment and / or control. A novel approach combining black-box and grey-box modeling techniques is integrated. Initially, a black-box fit is employed to generate model parameters based on training data specific to a building. These parameters are then used to establish initial parameters for a grey-box model through a parameter mapping process, which translates black-box model parameters into an equivalent set of grey-box model parameters while employing predefined equations and constraints to align the grey-box model with outcomes from the black-box model. The grey-box model parameters are further refined iteratively by fitting them against observed data using optimization techniques. This refinement aims to minimize discrepancies between model predictions and actual observations, with constraints imposed on the black-box model parameters to restrict the solution space during modeling.

This invention relates to the field of intelligent algorithm adversarial attacks, and in particular to a method and apparatus for black-box adversarial attacks on target detection models. The method includes: acquiring a white-box model with the same functionality as the black-box model; preprocessing the original samples and inputting them into the white-box model to obtain target detection boxes; initializing an adversarial patch and overlaying it onto the center position of the target detection boxes on each original sample to obtain adversarial samples; generating a variant of the white-box model; inputting the adversarial samples into the white-box model and the variant to obtain the detection results of the white-box model; constructing a loss function based on the game-like interaction of adversarial perturbation units; using the loss function to calculate the loss function value of the detection results; updating the adversarial patch according to the loss function value until the generated adversarial patch causes errors in the detection results of the white-box model; and utilizing the transferability of the adversarial patch to attack the black-box model. This improves the transferability of the adversarial patch and increases the success rate of attacks on the black-box model.

The application provides a method for determining a slip-thrust multi-stage structure recognition method, and relates to the technical field of geological structure judgment, and comprises the following steps: S1: building an experimental sand box model according to geological data; S2: changing the structure of the experimental sand box model to simulate a multi-stage structure superposition process, and obtaining an area change amount set; S3: processing the area change amount to obtain final dynamic result data; and S4: judging the slip type through the final dynamic data. The experimental sand box model can be built to simply and effectively simulate the multi-stage structure superposition process, and the area change amount of each stage of simulated stratum can be obtained. Through modeling processing, ADS processing and fitting correction processing of the area change amount obtained by the experimental sand box model simulation, the correctness of the model simulation can be effectively verified, the accuracy of the slip type judgment is greatly improved on the premise of simplifying the slip type judgment process.

The utility model relates to the technical field of tank models, and discloses a support display device of a tank model, which comprises a base, a display stand is fixedly mounted at the top of the base, a mounting groove is formed in the top of the display stand, a motor is fixedly mounted in the mounting groove, a cover plate is fixedly mounted at the top of the mounting groove, and the motor is fixedly mounted in the cover plate. An output shaft of the motor penetrates through the cover plate and is fixedly connected with a rotating disc, and a tank box model body is arranged at the top of the rotating disc. The supporting display device of the tank container model has a multi-dimensional display function and an omni-directional rotary display system, omni-directional observation is achieved, different observation requirements are met, the device is provided with a three-color-temperature LED lamp strip, spotlighting is conducted on key parts, shadow display is avoided through the shadowless lamp effect design, and through the arranged telescopic air cylinder and clamping plates, the supporting display device is convenient to use and high in practicability. The tank model can be stably fixed and cannot incline in the display process, the rotating speed of the rotating disc can be changed by adjusting the rotating speed of the motor, the display effect is improved, and operation is more convenient.

The invention discloses a virtual construction system based on BIM progress plan simulation, and the three-layer architecture comprises a data layer which is in butt joint with a DDE platform, a Prunus system and a PDM database through a standard interface; the technical support layer is built based on C # and NETCore technologies, and a data mapping engine and a business logic processor are arranged in the technical support layer; a service presentation layer, HTML, JS, CSS and VUE technology construction; the core function assembly comprises an LF space box model generator, a multi-dimensional data mapping engine, a four-view comparison analyzer and a dynamic simulation engine. The method has the beneficial effects that a qualitative leap is realized in the aspects of data integration and processing efficiency, and intelligent fusion of multi-source heterogeneous data is realized by establishing a standardized PBS-CBS-LF mapping rule base and an automatic data association engine; on the premise that progress semantics are reserved, the LF space box model enables Web end loading time to be effectively controlled, meanwhile, smooth interaction under the condition that the component scale is large is supported, and the performance dilemma of a traditional BIM system in a large scene is effectively solved.

The invention discloses an injection molding system for an expansion appliance. Comprising a model import program module, an undercut program module, a steel wire model generation program module, a base model generation program module, a fairing program module, an embedding box model generation program module, an injection molding hole program module, an expansion device model program module, a clamping groove generation program module and an output program module. By digitally designing the orthodontic device base plate model and the embedding box model matched with the orthodontic device base plate model, outputting the designed three-dimensional model, installing the steel wire and injecting the resin, the manufacturing of the expansion orthodontic device is completed, and the manufactured expansion orthodontic device is high in uniformity, smooth in surface, simple and efficient in manufacturing process and suitable for large-scale production.

The application provides a plug-and-play black box model watermark embedding method and device. The method comprises the following steps: step 1, generating a watermark detection network M_pty and a watermark verification sample X_v; and step 2, embedding the watermark detection network M_pty obtained in step 1 into a target network M_target which needs to be protected. The application realizes watermark embedding by inserting a trained exclusive watermark verification model into the target model, without fine-tuning the parameters of the target model, greatly guarantees the fidelity of the model, saves the training time and the training resource cost, has strong universality for models with different structures and different data sets, and can be applied to real scenes to process real tasks with complex DNN models.

The embodiment of the invention discloses an article recommendation black box model optimization method and device, equipment, a medium and a product. A specific embodiment of the method comprises the following steps: sampling article identifiers from an article identifier set to form a sampling set; executing an iteration step: determining actual response ranking information of each article identifier in the sampling set based on an article recommendation black box model; training the initial fitting model according to the article identifier set, the sampling set, the actual response ranking information and the initial model parameter information; through the updated initial model parameter information, screening an article identifier meeting a preset optimization condition from the article identifier set; adding the screened article identifiers to a sampling set; and in response to the fact that the execution frequency is equal to the preset iteration frequency, determining the at least one article identifier meeting the preset ranking condition as a recommended article identifier set. The implementation mode is related to big data, and the practicability of optimizing the article recommendation black box model and the accuracy of the article recommendation black box model are improved.

An IPTV EPG page compatibility method, system and electronic device, the method comprising: if there is no compatibility EPG code corresponding to the set-top box model in the database, the user end loads the EPG page based on the original EPG code, uploads the screenshot image to the server end after screenshot; the server end performs block recognition on the screenshot image; in response to the screenshot image being too large or too small, the size of the screenshot image is adjusted to be consistent with the target image, and the size adjustment information is determined; the key features are identified; if there is a difference between the actual key features and the corresponding target key features, the difference information is obtained, the compatibility EPG code is generated according to the size adjustment information, the difference information and the original EPG code, and the EPG page of the set-top box model is deployed based on the compatibility EPG code. The EPG page can be automatically compatible with various models of set-top boxes, effectively improving the applicability, reliability and efficiency of the EPG page compatibility, and greatly reducing the cost.

The invention relates to the technical field of chip modeling, and provides a chip Bin Model protection ring management method, which comprises the following steps: protection ring generation: based on a nearest principle, performing parameter copying on Bin Model with structural parameters that lmax is the same as lmin or wmax / Nfinmax is the same as wmin / Nfinmin, performing expansion generation on a maximum L column and a maximum W row, the Bin Model at the top right corner is generated by copying corresponding parameters of the original binning model Bin Model at the top right corner of the outermost ring based on the nearest principle; the protection rings are deleted, wherein after the protection rings on the outermost circle are removed, extended extension values set by a user are added to the parameters lmax, lmin, wmax and wmin or Nfinmax and Nfinmin of the remaining sub-box models Bin Model on the outermost circle; and automatic sorting: performing automatic sorting on the generated or adjusted Bin Model array, and supporting a sorting mode of combining a plurality of width dimensions W and length dimensions L. Simulation requirements can be met, performance stability, reliability and process adaptability of devices in chip Bin Model processing can be guaranteed, and modeling efficiency and precision are improved at the same time.

A method, system, and computer program product for an interpretable, feature-based post-hoc black box explainer for univariate time series forecasters are provided. The method receives a set of time series forecasting predictions. The set of time series forecasting predictions are generated from a set of black-box models trained with an initial data set. The method generates a set of features based on at least a portion of the initial data set. A set of surrogate models are trained based on the set of time series forecasting predictions and at least a portion of the set of features. A subset of surrogate models is selected. Based on the subset of surrogate models, the method generates one or more explanation outputs for time series forecasting predictions of the set of black-box models.

The invention relates to the technical field of operation monitoring, and particularly discloses a facility operation monitoring method and system based on a grey box model, and the method comprises the steps: analyzing a received operation monitoring demand, so as to obtain a corresponding grey box model and associated monitoring data; acquiring facility simulation data based on the grey box model and the associated monitoring data; based on the facility simulation data, the associated monitoring data and the operation monitoring demand, generating an operation monitoring result through a pre-trained facility agent; wherein the facility agent is obtained based on a grey box model and facility historical data training. Through the technical scheme provided by the invention, the actual operation condition of the technological process executed by the industrial facility can be accurately described by constructing the grey box model, and the grey box model is utilized to perform agent training and cue word optimization in the agent application process, so that the intelligent analysis capability of the agent can be greatly improved.

The present application relates to the technical field of machine learning, and discloses a method and system for evaluating the degree of alignment between a fine-grained decision-making logic of an artificial intelligence black-box model and human cognition. The method and system can evaluate the degree of alignment between a decision-making logic of a model and human cognition by analyzing the distribution of interactions modeled by the model. The implementation of the method and system comprises the following steps: providing an input sample; using a black-box model to predict the input sample to obtain a prediction result of the model; modeling the interactions between input units of the sample on the basis of the output of the black-box model, calculating the interaction strength of combinations formed between the input units, and expressing the output of the black-box model as an "interaction effect" between the combinations of the input units; and using interaction-based evaluation metrics to evaluate the degree of alignment between a decision-making logic of the model and human cognition. The advantage of the present invention is that: a quantification method for evaluating the degree of alignment between a fine-grained decision-making logic of an artificial intelligence black-box model and human cognition is provided. Compared with previous research, the present invention can better characterize the difference between a decision-making logic of a model and human cognition, thereby reflecting potential representation defects of the model.

The utility model provides an automatic V-method production line for shuttling type molding rings of extra-large parts, which relates to the technical field of V-method casting, solves the problem of low production efficiency, and comprises a production bus, the molding rings are arranged in the production bus, and the molding rings adopt an operation mode of multi-model online and alternate molding of an upper box model and a lower box model. The model can be replaced on line, the working efficiency is greatly improved, a manual operation station and a model replacement station of the production bus are both located on the periphery of a molding ring and are open environments, operation is convenient, and transfer of sand cores for the production bus and a passage of people flow are guaranteed. And the working procedures of model film coating, robot paint spraying, drying, modeling, drawing, pouring, cooling and the like are all automatically completed in the production line, so that the production efficiency is greatly improved.

The present application relates to the field of environmental science and technology, and more particularly to a method for retrieving methane sulfonic acid particles in marine aerosol based on a box model, comprising: step 1: obtaining satellite remote sensing data, ground observation data and laboratory simulation data, and performing data standardization processing; step 2: dividing the target sea area into multiple sub-regions as independent boxes; step 3: calculating the marine aerosol emission flux based on the concentration of dimethyl sulfide and wind speed, and updating the spatial and temporal distribution of the emission source; step 4: embedding the chemical reaction path of dimethyl sulfide oxidation to generate methane sulfonic acid on the surface of aerosol in the box model, and introducing the specific surface area of aerosol to correct the reaction rate; step 5: outputting the spatial and temporal distribution of methane sulfonic acid concentration and verifying it by comparing the measured data. The accuracy of aerosol source term estimation is improved, the fusion efficiency of multi-source data is improved, and the adaptability of the model in complex environment is enhanced.

Methods, apparatus, devices, and media for managing 3D models are provided. In one method, a panoramic image and a 3D model of a real scene to be presented in a virtual roaming application are acquired. The 3D model and a bounding box model of the 3D model are loaded at the viewpoint of the panoramic image. The bounding box model has an initial size larger than the size of the 3D model. The size of the bounding box model is adjusted based on the size of the 3D model such that the distance between the outer contour of the adjusted bounding box model and the outer contour of the 3D model satisfies a predetermined threshold condition. Using exemplary implementations of this disclosure, containing a bounding box model outside the 3D model can reduce undesirable visual impressions of holes.

The invention relates to the technical field of offshore oil pipeline laying simulation experiments, in particular to a full coupling effect testing device for S-shaped laying of pipelines, and aims to solve the problems that an existing experimental device is poor in real-time performance and cannot effectively simulate full coupling response of a pipe laying ship-stinger-pipeline-soil body system in S-shaped laying of the pipelines. The device comprises a control mechanism, an adjusting mechanism and a monitoring assembly. The control mechanism is fixedly connected with one end of the pipeline and can control the displacement direction of the pipeline, and the other end of the pipeline is hinged to the upper portion of the soil box model. The soil box model is arranged at the bottom of the test pool; the adjusting mechanism abuts against the side wall of the pipeline and can adjust the angle between the pipeline and the horizontal plane. The monitoring assembly can collect parameters of the pipeline in real time. The ocean ground is simulated through the soil box model, the control mechanism simulates the pipe laying ship, the adjusting mechanism changes the S line shape of the pipeline and the angle between the pipeline and the horizontal plane by adjusting the heights of different parts of the pipeline, and then various parameters of the pipeline can be collected in real time through the monitoring assembly.

Embodiments of the present application disclose a model processing method and device, a storage medium and an electronic device. A control model is obtained, the control model is provided with a plurality of control points; a bounding box model surrounding the control model is created, a target plane in the bounding box model facing a virtual camera is determined; the target plane is segmented based on current positions of the plurality of control points to obtain a segmented plane corresponding to each control point in the target plane; movement parameters required for each segmented plane to realize a target morphing motion special effect are determined, wherein the movement parameters of at least two segmented planes are different; the segmented planes are controlled to move according to the movement parameters to obtain a target model with the target morphing motion special effect. In this way, the target model with the target morphing motion special effect can be generated by simulating the target plane with a fission effect and moving the segmented planes in the target plane, the generation difficulty of the target morphing motion special effect is simplified, and the model processing efficiency is effectively improved.

The utility model discloses an octagonal box making machine which comprises a working platform used for installation and a first transmission rod fixedly installed at the upper end of the working platform. The device further comprises a first motor, the right end of the first transmission rod is fixedly connected with the output end of the first motor, a first bevel gear is installed at the left end of the first transmission rod, a second bevel gear is connected to the lower end of the first bevel gear, a threaded rod is installed in the second bevel gear, and a connecting block is connected to the outer side of the threaded rod. The outer side of the connecting block is connected with a first sleeve, the outer side of the first sleeve is provided with a second sleeve, the left end and the right end of the second sleeve are provided with sliding blocks, the left end and the right end of each sliding block are connected with a connecting rod, and the lower end of each connecting rod is connected with an octagonal box model. According to the octagonal box making machine, an octagonal box can be automatically and rapidly packaged conveniently, an octagonal box model needing to be packaged can be conveniently limited, and the formed octagonal box can be rapidly separated from the lower portion of the mold conveniently.

An object of the invention is to recommend more appropriate content to a user according to the user. An information processing apparatus includes: an acquisition unit that acquires input data indicating item characteristics; an item space learning unit that converts the vectorized input data into an N-dimensional rectangular region (Box) (where N is an integer of 2 or more); and a personal space learning unit that adjusts the Box model on the basis of user data of a specific user.