46 results about "Procedural modeling" patented technology

The invention provides a circuit board production yield root cause tracing method, which comprises the following steps of: acquiring process parameters, equipment states, environment variables and quality detection results of a whole production process, and constructing a multi-dimensional time sequence database; extracting a typical manufacturing process modeling unit through a sliding time window and dynamic time warping; establishing a cross-process dynamic causal relationship graph in combination with nonlinear Granger causal test, a structural equation model and a dynamic Bayesian network; an intervention and anti-factual reasoning method is applied, the causal effect and path stability under parameter disturbance of each process are evaluated, and the influence of a key causal path is quantified; according to the method, the accuracy of defect rate root cause positioning can be improved, and powerful support is provided for circuit board production process optimization and quality improvement.

The invention discloses a scene generation method and device based on multi-source GIS data fusion, and relates to the technical field of digital twinning and programmed generation. The method comprises the following steps: acquiring GIS data, preprocessing the GIS data, and storing the preprocessed GIS data in a geographic information resource library; a structured resource library is constructed, semantic parameters are added to the three-dimensional model in the structured resource library through the configuration file, and three-dimensional model resources with structured semantics are constructed; on the basis of the three-dimensional model resources with structured semantics and GIS data in a geographic information resource library, building and road generation and terrain processing are carried out in a programmed modeling engine through a configuration file, and scene data are generated; and importing the generated scene data into a real-time rendering engine, carrying out dynamic environment interaction and biocenosis simulation, and generating a city scene. The problems that in the prior art, an urban three-dimensional modeling method is low in efficiency and insufficient in environment interaction reality sense are solved.

The application discloses a time capsule expression structure reservation method based on deep learning and a knowledge graph, and aims at solving the problems that the expression and organization structure of a user's sealed multi-modal content in a digital time capsule service is difficult to be identified and recorded, and the display according to the uploading time or the fixed template of the media type in the opening stage is easy to cause the distortion of the expression intention and the fragmentation of the receiving experience. The multi-modal content input in the sealing stage is normalized and time-aligned, the content representation is obtained by using multi-modal hierarchical coding, the segmentation boundary and hierarchical relationship are determined and the key weight and object information are extracted by combining span boundary detection and a pointer network, the narrative rhythm parameters are obtained by using a neural time point process modeling based on a media switching event sequence, a timing knowledge graph is fused and constructed, and the arrangement sequence and the presentation control parameters are generated under the sequence constraint and the segmentation continuous constraint by graph representation learning, and the presentation data for the display in the opening stage is generated, so that the technical effects of reserving the user's expression structure and rhythm and differentially arranging and presenting in the future are realized.

This invention relates to a Text2SQL self-correction method based on error pattern awareness, belonging to the field of data processing technology. It includes: abstracting general error patterns from isolated error messages to enable the model to have generalizable diagnostic capabilities, and constructing a knowledge graph based on these error patterns; abandoning traditional single-step blind correction, constructing a dynamic feedback loop containing state memory to force the model to escape local optima; and performing adaptive iterative repair termination based on a comprehensive score. This invention models the SQL correction process as a closed-loop control flow of "diagnosis-retrieval-verification," breaking through the traditional "static verification + manual rules" paradigm. It transforms database execution feedback into a structured error pattern representation, guiding a Large Language Model (LLM) to establish generalizable SQL error diagnosis and repair capabilities.

The invention provides a machining process modeling method based on dimension reduction coding. The machining process modeling method comprises the steps of 1, obtaining discrete and analyzable machining paragraphs in a complex machining process; step 2, establishing a relative coordinate system based on a processing section corresponding to each section of processing after discretization in the complex processing process, and expressing each section of processing under the relative coordinate system; 3, calculating the vector difference of the local features of each processing section in the corresponding complex processing process under the relative coordinate system; and 4, according to each processing section, establishing a serialized expression of the processing process based on the vector difference and the processing parameters. According to the method, the vector difference under the relative coordinate system is adopted to reflect geometric or physical changes between complex machining sections, local features of complex machining are highlighted, different machining conditions are adapted through local changes, and the universality of the model is enhanced.

The application discloses a kind of broken screening process circulation steady-state solving method, system, medium and computer program product, belong to industrial process modeling and numerical solving technical field, comprising: obtaining the process configuration data of broken screening process;Directional graph model is constructed based on process configuration data;Establish equipment operating condition calculation model for each node;Cyclic structure identification is executed to directed graph;Linear pre-computation is executed to linear region, and the external input boundary of each cyclic subgraph is obtained;According to solving level, each cyclic subgraph is executed iteration solving, and the node output data of this round is obtained;In iteration solving process, iteration stabilization processing is executed;After reaching the minimum iteration number of pre-set, based on the error between the adjacent iteration output data of each node in cyclic subgraph, all node convergence determination is executed, when the error of all participating determination nodes meets pre-set convergence condition, determine that cyclic subgraph reaches steady state;Steady-state calculation result is output.

The application discloses a control-oriented sintering endpoint modeling method based on a coupling mapping framework, and belongs to the technical field of metallurgical industrial process modeling and process control. First, sintering endpoint time series data and multiple sintering data time series sequences are collected, and a sintering endpoint modeling model containing an attention gate cycle unit module and a bidirectional gate cycle unit is constructed; the sintering data is divided into manipulated variables, coupling disturbance variables and disturbance variables; after the various time series data are divided into time series segments, the coupling disturbance variables and the manipulated variables are respectively subjected to the attention gate cycle unit module to obtain state vector prediction values, and then the state vector prediction values are input into the bidirectional gate cycle unit together with the divided manipulated variables, disturbance variables and coupling disturbance variables to obtain sintering endpoint single-step prediction values; further, a loss function is calculated according to true values, and sintering endpoint modeling model parameters are iteratively updated. The method guarantees a simple model structure, and takes into account prediction accuracy and control availability.

The invention relates to the technical field of artificial intelligence and industrial process modeling, in particular to a time series data generation method and device based on a Gamma variational auto-encoder, and the method comprises the steps: carrying out the modeling of potential variables in the variational auto-encoder through Gamma distribution, and carrying out the modeling of the potential variables when the variational auto-encoder is trained, firstly, each variable value of original time sequence data in a training set is normalized according to a constraint condition to serve as training input, and then a micro-weight parameterization sampling mechanism is introduced to sample Gamma distribution constructed by encoder output parameters, so that the problem that training is unstable under the condition of small shape parameters in a traditional sampling method is solved. Therefore, the trained variational auto-encoder can significantly improve the constraint satisfaction rate and physical consistency of the generated time sequence data. According to the method, the high-quality time sequence data with good engineering availability can be stably generated.

This invention relates to a quality evaluation and automatic process control technology for PCBA welding scenarios, comprising: establishing a quantitative standard evaluation system for welding quality based on multiple performance indicators of the product; fusing multimodal sensor data with process parameters to construct a mathematical model; and optimizing processing parameters based on machine learning. Starting from the intelligent problem of PCBA welding process quality control, this invention proposes a quantitative evaluation system for PCBA welding quality, uses process modeling based on multi-sensor fusion of testing, innovatively introduces machine learning-based process parameter optimization methods, develops a PCBA welding prototype with online process control, achieves closed-loop automatic process optimization, and simultaneously solves the crucial data collection problem in big data analysis.

The embodiment of the present application relates to the technical field of computer-aided software engineering, and discloses a process modeling data processing method for human-computer collaboration and related equipment, which comprises: establishing a double-storage architecture for process modeling, including a business data storage module and a layout data storage module, the business data storage module storing entity data, and the layout data storage module storing relative coordinates of each node in a swimlane relative coordinate system and arrangement order of the swimlane on a canvas; in response to an adjustment operation of the swimlane arrangement order, obtaining the adjusted swimlane arrangement order and storing it in the layout data storage module; determining the offset of each swimlane on the canvas based on the adjusted swimlane arrangement order during rendering; and determining new absolute coordinates of the nodes in the canvas according to the relative coordinates and the offset of each swimlane and performing rendering update. Through the above method, the embodiment of the present application decouples business data and layout data, can improve modeling efficiency, reduce overhead and significantly improve performance.

The invention provides a hardened cement paste multi-scale direct coupling calculation method and device, and the method comprises the steps: carrying out the in-situ loading and computed tomography of hardened cement paste, generating multi-scale volume data of different stages, and carrying out the multi-phase segmentation to obtain a multi-valued volume data mask; the damage path prediction deep learning model is used to predict the damage path of the multi-valued volume data mask and divide a multi-scale grid, the finite element software is used to carry out direct simulation calculation on the multi-scale grid structure to generate a multi-scale damage result, and the high information density data obtained based on the image can be used to predict the damage path of the multi-valued volume data mask. A potential damage path is predicted through a deep learning model, modeling is performed on the basis, calculation complexity is simplified, on the premise that simulation precision is guaranteed, microstructure evolution characteristics and macroscopic mechanical property response of cement paste in the hydration process are considered, the numerical calculation cost of a multi-scale structure is effectively reduced, and the method is suitable for popularization and application. And an efficient and extensible technical path is provided for evolution process modeling of complex multi-phase materials.

The invention discloses a data sample screening and reconstruction method in high nickel matte flotation process modeling, relates to the technical field of high nickel matte flotation process data modeling, and solves the technical problem of data validity in an actual modeling process. The method comprises the following steps: selecting a system lag parameter p according to the actual process of a high nickel matte flotation system, and reconstructing data samples; marking invalid data; integrating discontinuous data in time, and constructing mutually independent sample data sets; carrying out matrix recombination on the extracted samples, and carrying out RNN modeling. According to the invention, data are preprocessed according to the actual working condition of high nickel matte, invalid data are removed, and some data blocks which are discontinuous in time are obtained.

This application relates to a method for allocating interference resources for a drone swarm based on a pre-trained attention encoder. The method includes: setting an objective function for the interference resource allocation problem using an interference effect evaluation model and the interference power consumed by the drone swarm; modeling the interference resource allocation problem as a constrained combinatorial optimization problem by considering the limitations on drone communication, energy, and flight speed under denial conditions; modeling the interference decision-making process in the combinatorial optimization problem as a distributed locally observable Markov game model; mapping drones as agents; and solving the model using the MAPPO algorithm based on a pre-trained attention encoder to obtain the interference resource allocation scheme for the drone swarm. This method can solve the problems of state space dimensionality explosion and environmental non-stationarity in multi-agent collaborative decision-making processes.

The invention discloses a process modeling and fault detection method and system for machine learning guided by a control theory, and the method comprises the steps: obtaining historical operation data of an industrial process, and extracting an operation dynamic state in the historical operation data through a neural network, approximation is carried out on an observer gain part of the kernel representation model to realize complete kernel representation model design; building an adjoint system of the nuclear representation model by using a Hamilton system theory, and performing guided optimization on the learning and training process of the nuclear representation model and the adjoint system thereof by using nondestructive constraint to realize the design of a regularized nuclear representation model; and a detection statistic is constructed by using the residual error of the kernel representation model and the regularization kernel representation model, and the maximum value of the detection statistic during normal operation of the industrial process is set as a fault detection threshold, so that fault detection of the industrial process is realized, and the safety of the industrial process is guaranteed. In addition, by designing an online fine tuning strategy, generalization and applicability of the model are improved, and the method has high practical value.

The embodiments described herein include a process for generating a hybrid model for modeling a process in a semiconductor processing equipment. In a particular embodiment, a method for creating a hybrid machine learning model includes identifying a first set of cases spanning a first range of process and / or hardware parameters and conducting laboratory experiments for the first set of cases. The method may further include collecting experimental outputs from the experiments and performing physics-based simulations for the first set of cases. In one embodiment, the method may further include collecting model outputs from the simulation and correlating the model outputs with the experimental outputs using a machine learning algorithm to provide a hybrid machine learning model.

The invention provides a multi-modal learning data acquisition and standardization method based on a smart writing system, and relates to the technical field of data processing. Disturbance signals are acquired through multi-modal sequences such as pressure, speed, stroke continuity, sight line coordinates and voice energy; nodes are generated in a node type mapping table according to the disturbance amplitude interval number, the change rate interval number and the duration interval number; calculating a disturbance amplitude difference, a change rate difference and a node time interval in a preset time window to form an edge parameter; compressing the node set and the edge set through the node difference degree and the edge parameter difference degree; and performing direction verification and parameter truncation on the structure according to the topology constraint table, and finally outputting a standardized causal graph structure comprising a node list area, an edge list area and a topology description area. According to the method, output structure consistency can be kept under the condition that the number of multiple modes is inconsistent or partial modes are missing, and the method is suitable for scenes such as learning behavior collection, writing process modeling and educational data processing.

The present application relates to the technical field of artificial intelligence and industrial process modeling, in particular to a time series data generation method and device based on Gamma variational autoencoder, comprising: modeling the latent variables by using Gamma distribution in the variational autoencoder, and when training the variational autoencoder, first, according to the constraint condition, normalizing each variable value of the original time series data in the training set as the training input, and then introducing a differentiable reparameterization sampling mechanism to sample the Gamma distribution constructed by the encoder output parameter, so as to solve the training instability problem existing in the traditional sampling method under the condition of small shape parameter, so that the trained variational autoencoder can significantly improve the constraint satisfaction rate and physical consistency of the generated time series data. The present application can stably generate high-quality time series data with good engineering usability.

The embodiment of the application relates to the technical field of software engineering and artificial intelligence, and discloses a process modeling method for human-machine cooperation and related equipment, the method comprising: constructing a hierarchical recursive process model, the process model comprising a process domain layer, an activity layer and a task layer; using an eight-tuple model to formally describe each task, the eight-tuple model comprising a task identifier, an input artifact set, an execution logic, an output artifact set, a role performer, a constraint set, a cognitive level and a deterministic classification, defining four types of role performers for the process model, including a human role, an agent role, a system role and a hybrid role, and based on the four types of role performers and the eight-tuple model, rendering each activity of the process model as an activity graph on a preset workbench interface through differentiated visual lanes, and in this way, the embodiment of the application effectively realizes modeling of AI tasks under the framework of organizational system constraints and standard specifications in a human-machine cooperation scenario.

The invention discloses a document-level event argument extraction method and device, and the method comprises the steps: building a table structure representation based on an event trigger word and a predefined role, and carrying out the hierarchical coding of a document, so as to obtain the cell embedding based on the event trigger word, the role and the argument; attention bias guided by argument dependency is introduced in the layered coding process, and the argument dependency type comprises intra-event dependency and inter-event dependency. Dynamically setting a plurality of role prototypes for each role in the hyperspherical space, modeling the role prototypes and the argument allocation process as an optimal transmission problem, constructing optimal transmission loss, and constructing prototype optimization loss in combination with prototype separation regular term loss; constructing a role graph between the events, spreading role semantics in a document based on a graph neural network, and constraining the consistency of the same roles between the events by adopting contrast loss; and training the pre-training language model by using the total loss formed by the argument extraction loss based on the range, the prototype optimization loss and the comparison loss, and outputting the argument fragment corresponding to each role.

This invention discloses an anti-jamming zero-sum Markov game model and a maximum-minimum deep Q-learning method. The model considers a communication adversarial scenario involving a pair of transceivers, a fixed jammer, and an intelligent jammer. The fixed jammer releases frequency sweeping interference, the intelligent jammer optimizes the jamming channel selection, and the user optimizes channel and power selection to maximize transmission utility. The adversarial interaction process is modeled as an anti-jamming zero-sum Markov game. The method involves: initializing the anti-jamming network parameters and training hyperparameters; constructing a spectrum waterfall plot and inputting it into the anti-jamming network, outputting an adversarial Q-value matrix; calculating and executing anti-jamming actions, calculating the reward value obtained for the current action and saving the adversarial record; sampling the adversarial record, calculating the Q-value estimation error, and updating the anti-jamming network using a backpropagation gradient algorithm; repeating the above interaction process until network training is complete. This invention effectively solves the problem of non-stationary environmental changes caused by jamming strategy updates, obtaining a more robust anti-jamming strategy.

The application discloses a kind of execution chain segmentation type construction methods for heterogeneous computing platform, it is related to heterogeneous computing and real-time task management technical field.The present application takes the end-to-end execution flow of complete complex application as object, by identifying the basic execution unit in original application execution flow, the execution resource type and effective interaction boundary of each basic execution unit are combined, and the execution chain model composed of multiple execution segments is constructed, and the execution chain model obtained can be directly served for subsequent scheduling analysis, execution control, timing analysis and cross-platform deployment.The present application can accurately describe the real execution process of complex intelligent application on heterogeneous platform, significantly improve the expression ability of execution chain model to end-to-end application flow, improve the integrity and accuracy of heterogeneous execution process modeling, with strong engineering practicability and industrial application prospect.

The application discloses a device for supporting complex system requirement integrity modeling and evaluation, and relates to the technical field of system engineering, and comprises requirement modeling, requirement integrity evaluation, design process modeling, requirement and design process ontology integration, requirement traceability expression and evaluation; the requirement modeling module comprises requirement entry modeling and requirement content constraint definition; the requirement modeling module mainly functions to realize requirement model creation based on requirement integrity expression rules and constraints, and guarantee requirement content integrity in the created requirement model; in the requirement modeling module, the requirement content constraint definition submodule provides constraints to the requirement entry modeling submodule; the application uses entry modeling technology to decompose high-level requirements into specific and operable entries in the requirement modeling module, which promotes in-depth understanding of the requirements and ensures that each requirement is accurately defined and recorded.

An apparatus and method for multiple stage process modeling is provided. The apparatus includes a processor and a memory connected to the processor. The memory containing instructions configuring the a processor to receive process data sets, each process data set representing a progression stage that describes a sequence of activities performed by an entity device, generate, using the process data sets and a machine learning algorithm, a progression outlook profile including progression stage profiles, each progression stage profile representative of a respective progression stage and may generate progression actions describing progression from a first progression stage to a second progression stage based on input data, and a progression stage profile classifier that may use input data and identify a progression stage currently occupied by a process based on input data. The processor may receive process data describing a process to classify received process data to a progression stage profile.

This invention relates to an autonomous driving scene reconstruction method based on MCMC and scale-aware time-varying scheduling, comprising: initially modeling the scene as a set of anisotropic 3D Gaussian spheres, each Gaussian sphere being parameterized by learnable parameters to construct a learnable parameter space; iteratively reconstructing the autonomous driving scene based on the learnable parameter space, wherein in each iteration, a scale-based semantic masking mechanism is used to automatically distinguish scene regions using the statistical distribution characteristics of the Gaussian sphere scale; wherein, during the iteration process, based on the stochastic gradient Langevin dynamics update mechanism, an adaptive time-varying noise scheduling strategy is introduced to model the noise injection process in Markov chain Monte Carlo MCMC sampling as a control process that dynamically changes with the number of training iterations. Compared with existing technologies, this invention has the advantages of stable exploration capability and robust geometric constraints.

The application discloses an artistic design assisting method and system based on artificial intelligence, relates to the technical field of artificial intelligence and computer aided design, and comprises the following steps: constructing an artistic design logic structure diagram to perform coding and mapping, generating a latent vector, preprocessing and performing semantic extraction on target design reference information input by a user to generate a target embedding vector; performing dimension reduction on the latent vector through Householder reflection, performing zero padding, inverse mapping and decoding reconstruction on a low-dimensional Bayesian optimization vector to obtain a candidate image and acquire a candidate embedding vector, calculating a target consistency loss value in combination with the target embedding vector, converting the target consistency loss value into a maximum score function, performing Gaussian process modeling and sampling, generating an optimal result, constructing a pair-wise preference probability function, constructing a ranking likelihood function to perform parameter updating and iteration, and obtaining a final image; and the application makes the artistic design assisting process more adaptive and flexible.

Described are 3D printing platforms comprising stereolithographic 3D printing devices utilizing a static optical assembly and procedural modeling applications representing 3D scenes as signed distance function. Described are also structures such as bioreactors that can be printed using such platforms, as well as characteristics and used thereof.

The invention aims to provide an assembly line variant path planning system and method based on graph depth reinforcement learning. The system comprises an input layer, a knowledge and mechanism layer, a core processing module, an output layer and an application and feedback layer. According to the method, the assembly line deformation process is modeled into a graph structure, and a reward function accurately matched with a change propagation optimal target is designed. An input change propagation optimal mathematical model is used as a training basis, and a neural network model constructed based on graph depth reinforcement learning is repeatedly trained until a model output result is stably converged, so that an optimal preliminary change path can be intelligently identified. The optimal path provides a key dimension linkage framework for subsequent instantiation reconstruction, virtual-real mapping rapid modeling of the whole assembly process is finally achieved, and the reverse modeling time of the physical assembly line is remarkably shortened.

The embodiment of the present application relates to the technical field of computer-aided software engineering, and discloses a process modeling data processing method for human-computer collaboration and related equipment, which comprises: establishing a double-storage architecture for process modeling, including a business data storage module and a layout data storage module, the business data storage module storing entity data, and the layout data storage module storing relative coordinates of each node in a swimlane relative coordinate system and arrangement order of the swimlane on a canvas; in response to an adjustment operation of the swimlane arrangement order, obtaining the adjusted swimlane arrangement order and storing it in the layout data storage module; determining the offset of each swimlane on the canvas based on the adjusted swimlane arrangement order during rendering; and determining new absolute coordinates of the nodes in the canvas according to the relative coordinates and the offset of each swimlane and performing rendering update. Through the above method, the embodiment of the present application decouples business data and layout data, can improve modeling efficiency, reduce overhead and significantly improve performance.

Devices, methods, and systems for process modeling using tuning factors are described herein. One method includes receiving, by a computing device, a model representing a physical process, wherein the physical process involves a plurality of equipment items and generates a physically measurable output variable, determining, by the computing device, a predicted value of the physically measurable output variable using the model, and adjusting, in the model, a tuning factor associated with an equipment item of the plurality of equipment items responsive to a difference between the predicted value of the output variable and a physically measured value of the output variable exceeding a difference threshold.

The invention relates to an unmanned surface vessel game confrontation control method and device based on LW-PPO, a program and a storage medium, and belongs to the technical field of USV intelligent control and deep reinforcement learning. The method comprises the following steps: performing Markov game process modeling on a game environment and initializing a role state; acquiring a previous moment hidden state vector and a current moment state vector, and inputting the current moment hidden state vector and the current moment state vector into a liquid neural network for time sequence feature extraction to obtain a current moment hidden state vector; inputting the vector into a strategy network to generate a selected action; the selected action is converted into acceleration and steering rate control quantity through a PID controller; performing kinematics updating according to the control quantity and the current state to obtain an updated state vector; interacting with the environment to judge whether to end, and if not, circularly executing; liquid neural network explicit modeling USV continuous time dynamics is introduced, and the time sequence feature extraction capability is enhanced; and traditional KL divergence constraint is replaced by a smooth regular term, so that the strategy updating stability is improved.