28 results about "Static model" patented technology

This invention relates to the field of new energy technology and discloses a short-term wind speed prediction method based on LSTM neural networks. The method first collects multi-dimensional historical wind speed data from wind farms and performs standardized preprocessing to construct a high-quality training set. Then, it designs and trains an LSTM neural network model that can effectively capture temporal dependencies. Finally, it obtains a high-precision prediction model through validation and optimization. In application, the preprocessed real-time data is input into the model to achieve quantitative prediction of wind speed for the next few hours. This method establishes an online performance evaluation and automatic update triggering mechanism based on the inherent volatility of the data, forming a complete prediction-evaluation-optimization closed loop. This allows the model to self-perceive performance degradation and autonomously trigger retraining, thereby overcoming the problem of performance degradation of traditional static models due to environmental changes after long-term deployment. Ultimately, this ensures the accuracy, adaptability, and industrial application value of the prediction model in long-term operation.

This invention relates to the field of reservoir engineering management technology, specifically a reservoir lifecycle management system based on spatiotemporal matrix and digital twin. The system includes a digital twin model construction module, a spatiotemporal matrix module, a data fusion module, a lifecycle simulation module, a real-time data acquisition and update module, a model dynamic adjustment module, and a management decision support module. The system uses a spatiotemporal matrix to uniformly organize and manage multi-source data throughout the reservoir's lifecycle, utilizes a digital twin model to construct a virtual representation of the reservoir, achieves bidirectional synchronization through data fusion, and dynamically adjusts model parameters based on real-time data, ultimately generating lifecycle management decision outputs. This invention solves the problems of scattered data and static models in existing technologies, improving the accuracy and reliability of reservoir management.

This invention belongs to the interdisciplinary field of artificial intelligence and geological resource assessment, specifically relating to a method and system for predicting the energy storage potential of salt caverns based on multi-source geological information. It aims to address the problems of low prediction accuracy and difficulty in balancing breadth and precision in existing technologies due to single-source data, static models, and lack of adaptive capabilities. The method includes: acquiring multi-source data such as seismic, well logging, core, and geostress data; constructing a geological feature fusion vector containing seven core parameters; establishing a high-dimensional nonlinear mapping model driven by a deep neural network; dynamically adjusting the prediction range based on data coverage density and confidence level; and achieving continuous model iteration through online incremental learning. The system integrates data quality assessment, anomaly handling, and distributed computing modules. By adopting the above technical solutions, this application can achieve a significant improvement in prediction accuracy and adaptive intelligent adjustment of the prediction range.

A method and apparatus for predicting the configuration and attitude of a towed wire array subjected to current-induced vibrations, relating to the field of marine exploration technology, are disclosed. The method includes: employing a quasi-static model to determine the configuration and attitude and tension of the towed wire array at its equilibrium position by solving a set of nonlinear equations considering boundary constraints; wherein the tail boundary conditions of the quasi-static model include tail rope tension; employing a dynamic model, using the configuration and attitude and tension as initial conditions, discretizing and solving the dynamic model to obtain the spatiotemporal distribution characteristics of the displacement of the towed wire array subjected to current-induced vibrations; wherein the dynamic model includes segment tension and gravity. This application combines a quasi-static model and a dynamic model, adding tail rope tension, segment tension, and gravity to the existing parameters, making the model conform to the actual stress conditions of the towed wire array, thereby achieving accurate prediction of the spatiotemporal distribution of the configuration and attitude of the towed wire array subjected to current-induced vibrations.

This invention provides a MetroAnalyzer software system that integrates CD-SEM and GDSII / OASIS for layout measurement and analysis. Relating to the field of data processing technology, it deeply integrates quantum computing and AI technologies. Through quantum annealing optimization algorithms and multi-dimensional feature embedding methods, it achieves ultra-efficient deviation prediction and measurement optimization. Utilizing a quantum computing processor, it decomposes multi-MP measurement tasks into quantum bit states, reducing repetition rate and shortening detection time. Simultaneously, the AI ​​multi-dimensional feature embedding method, combined with temporal convolutional networks and a Transformer architecture, extracts spatial-temporal features from historical data. An adaptive knowledge distillation mechanism controls prediction errors to within 0.5nm, avoiding the limitation of traditional static models with errors exceeding 1nm, significantly improving measurement efficiency and prediction accuracy. Furthermore, through a quantum-classical hybrid architecture, it achieves seamless integration from deviation identification to optimization, breaking through the bottleneck of traditional reliance on manual intervention and repeated experiments, providing a novel solution for high-precision, high-throughput semiconductor manufacturing.

The application relates to the technical field of industrial control, in particular to a control method and system of an insulating material production line. The method comprises the following steps: acquiring initial operation data of the insulating material production line in real time; calculating a transient specific energy index in combination with a standard plasticization reference absolute temperature; extracting a pressure change rate by performing smooth filtering on a time differential item of a head melt pressure; deducing a dynamic thinning coefficient in combination with the transient specific energy index, a value at a previous sampling moment, the pressure change rate and the head melt pressure; constructing a state matrix mismatch correction factor based on the dynamic thinning coefficient, a historical maximum thinning limit value of a sliding data window and a standard steady-state thinning reference value; and performing adaptive model predictive control to output an optimal control instruction by using the correction factor to target correct a discrete state space model system matrix. The application can effectively solve the mismatch problem of a static model under a nonlinear mutation condition, improve control precision and reliability, and guarantee stable and efficient operation of the production line.

The application provides a tunnel face information-based dynamic geological model updating and predicting method, which combines the tunnel face grid point data obtained in the tunnel excavation process, adopts Kriging interpolation and variogram fitting technology, and constructs and dynamically updates the geological model. The method comprises the following steps: collecting the face data, cleaning and standardizing the data, and extracting the tunnel starting end face and the ending end face; generating an initial geological model by using a three-dimensional variogram model and Kriging interpolation; integrating the newly added face data, dynamically adjusting the model parameters, and updating the geological distribution; and finally displaying the dynamically updated geological model through three-dimensional visualization technology. The application solves the problem that the traditional static model is difficult to adapt to the dynamic changes of geological conditions in the excavation process, can optimize the geological model in real time, provides accurate and reliable geological information support for tunnel construction, and significantly improves the construction efficiency and safety.

The present application belongs to the technical field of slope dynamic monitoring, and discloses a kind of open pit mine slope dynamic monitoring and stability evaluation system and method.The method includes the spatiotemporal alignment and dynamic weighted fusion of multi-source heterogeneous data;According to the results of spatiotemporal alignment and dynamic weighted fusion of multi-source heterogeneous data of microseismic-stress coupling damage factor, rock mass damage factor real-time inversion based on microseismic energy density is carried out, and rock mass damage error is eliminated through spatiotemporal-Bayesian cascade weighting;Based on the dynamic stability coefficient of improved coupling rock mass damage factor and hydrological parameter, the dynamic evaluation of water-rock coupling stability is carried out;Case base driven adaptive early warning threshold generation.The present application greatly improves the success rate of its own prediction and forecast;The damage inversion model compresses the stability coefficient error to within 5%, which is 3 times higher than the static model.

This invention discloses an information processing method and system, relating to the field of information processing technology. It includes S1, acquiring multi-source heterogeneous data and adding multi-dimensional time-series labels to the multi-source heterogeneous data; S2, constructing a dynamic association learning model, dynamically generating and updating a data association rule base through real-time collected features of the multi-source heterogeneous data, introducing a time-series attention mechanism to obtain dynamic association rules. This invention provides accurate time dimension basis for data processing by acquiring multi-source heterogeneous data and adding multi-dimensional time-series labels containing collection timestamps, transmission delay compensation values, and time sensitivity levels. Combined with the construction of a dynamic association learning model incorporating a time-series attention mechanism, it collects the frequency of associations and collaborative change trends of data fields in real time to dynamically update the association rule base, effectively capturing dynamic coupling relationships between data, and solving the problem that existing technologies relying on static models cannot adaptively identify implicit associations in multi-source heterogeneous data with strong time-series dependencies.

The application provides a bridge construction scene data modeling system and method based on space-time objects; the system comprises a data classification module, a framework modeling module, a multi-dimensional description module and a model integration module; the application introduces dimensions such as cognitive intelligence, behavior action and correlation relationship, so that the digital object has the ability of perception, analysis and decision-making, and realizes evolution from a static model to a dynamic intelligent entity; through systematic description of the composition structure and the correlation relationship, a knowledge network between objects is constructed, and a leap from data integration to semantic correlation is realized; through fusion of real-time perception and cognitive intelligence, the system can actively predict risks and optimize construction schemes based on environmental changes and object states, realizes self-adaptation and self-optimization of the construction process, and improves the intelligent management level of the project.

This invention relates to the field of power system planning and operation optimization technology, specifically disclosing a collaborative optimization method and system for multiple types of active support equipment at the weak-sending end. The method includes establishing a voltage-reactive power droop characteristic equation and constructing a unified power flow-droop static model to describe the system's steady-state operating point; constructing comprehensive performance indicators for multiple scenarios and a unified stiffness constraint model considering the short-circuit capacity of the external grid and the reactive power support capacity of the multiple types of active support equipment; at the capacity configuration layer, minimizing the weighted difference between investment cost and stability benefits is the objective; at the operation layer, minimizing the weighted sum of reactive power output adjustments for multiple types of equipment is the objective. This invention achieves coordinated improvement in configuration and operation to enhance the static margin in the most unfavorable direction and expand the stability domain form by solving capacity configuration using a surrogate model and linear programming at the upper layer, and optimizing reactive power sharing based on Q / V sensitivity and action costs at the lower layer.

The application provides a personalized model generation method, device and equipment based on multi-source feature cooperation and a medium. The method comprises the following steps: collecting a motion video stream of a target object based on a camera terminal; outputting a motion quantization index corresponding to the motion video stream based on a pose estimation model; acquiring external sensor data synchronized with the motion video stream; determining a prediction deviation vector between the motion quantization index and the external sensor data in response to a preset calibration trigger condition; and performing parameter compensation on the pose estimation model in the camera terminal by using an incremental learning algorithm according to the prediction deviation vector, so as to generate a personalized model adapted to the target object. The application gradually fits the personalized motion mode and physiological characteristics of the user, breaks through the parameter solidification bottleneck of the traditional static model, transfers the external sensor data to the visual model in a cross-modal manner, makes up for the limitations of single visual perception, realizes the deep fusion of multi-source heterogeneous data, and improves the reasoning accuracy and environmental adaptability of the model.

This invention discloses a method and apparatus for risk decision-making and governance of projects under construction based on reinforcement learning and dynamic game theory. The method includes: constructing an engineering game state space based on multi-source heterogeneous data; inferring the implicit payoff functions of participating agents based on maximum entropy inverse reinforcement learning; constructing a dynamic game adversarial training environment including a strategic data fraud mechanism; generating robust governance strategies based on meta-game theory and multi-agent reinforcement learning; performing proactive detection and dynamic risk locking based on maximizing information gain; and deploying and implementing closed-loop evolution of the governance strategies. This invention can identify and quantify the motivation for strategic data fraud, and the generated governance strategies have the ability to predict the opponent's reactions. It overcomes the problem of static models failing during adversarial data drift and can proactively break information asymmetry, solving the technical problem of risk governance failure caused by human game theory.

The application relates to a high-voltage frequency converter digital twin modeling system and method, and belongs to the technical field of digital twin equipment and methods. The technical scheme of the application is that a data acquisition module and a data preprocessing module are connected with each other, and a digital twin modeling module is connected with the data preprocessing module and a component degradation modeling module respectively. The application has the beneficial effects that by constructing a multi-physical-field coupling twin model, the problem of large simulation deviation of a single electrical model under non-ideal working conditions is solved, and the virtual-to-real mapping precision is greatly improved; by embedding a degradation model with a harmonic aging factor, the dynamic updating of model parameters with the equipment state is realized, and the problem of continuous expansion of the deviation of a static model with the running time is solved; by building a health-driven low-harmonic collaborative closed-loop control mechanism, the operation and maintenance and harmonic control are deeply integrated, and the practical value of the system engineering is significantly improved.

The application belongs to the field of bioinformatics and computational biology, and particularly relates to a mutation detection method and system based on an attention mechanism and dynamic routing. The method comprises: obtaining a sequencing sequence alignment result, and extracting candidate mutation sites with high confidence through threshold setting preliminary screening; a local sequence window is intercepted with the candidate site as the center, a multi-dimensional feature tensor is constructed by fusing physical space position, positive and negative chain alignment classification and sequencing quality attributes; the multi-dimensional feature tensor is input into a backbone network, background noise is suppressed by using a Gaussian guided space attention mechanism, and long-distance multi-scale mutation features are extracted by using a shift window mechanism of a Swin Transformer and a dynamic routing strategy of a mixed expert module; finally, the existence of mutations, zygote states and allele sequences are determined in sequence by using a multi-task dynamic decoding network, and network parameter optimization and result output are completed by using a cost-sensitive focal loss. The application avoids the statistical feature loss problem caused by traditional image mapping, alleviates the limitation of limited local receptive field, realizes adaptive adjustment of the calculation path according to the sequence complexity by introducing a dynamic routing mechanism, overcomes the configuration imbalance of static models in the allocation of computing power and the capacity of local parameters, and enhances the detection accuracy and stability of the model under complex variation and low sequencing depth conditions.

Embodiments of the present application provide a simulation model building method and a wafer scratch simulation method. The simulation model building method comprises: acquiring polishing parameters of a chemical mechanical polishing process that causes scratches, the polishing parameters including a first motion parameter and a second motion parameter, the first motion parameter being used for representing the motion trajectory of a polishing pad, and the second motion parameter being used for representing the motion trajectory of a wafer; and on the basis of the polishing parameters, establishing a static model and a dynamic model that are used for fitting the scratches, wherein the static model is configured to fit scratches caused by abrasive particles of which the positions on the polishing pad are regarded as unchanged, and the dynamic model is configured to fit scratches caused by abrasive particles of which the positions on the polishing pad are regarded as changing over polishing time. A static model and a dynamic model that are used for fitting scratches are established on the basis of polishing parameters, so that the accuracy of scratch fitting in the chemical mechanical polishing process can be improved, and the application range of scratch fitting is expanded, thereby improving the fitting effect of scratch fitting in the polishing process.

This invention discloses a friction feedforward compensation method and system for a servo system, comprising: identifying segmented friction force model parameters, including maximum static friction torque, Coulomb friction torque, viscous friction coefficient, and critical Stribeck speed, based on speed feedback signals and utilizing the steady-state response of the servo system to simple input signals; modeling the friction torque of the servo system as a piecewise function of speed according to the model parameters, constructing a segmented friction force model to calculate friction force; using the segmented friction force model as a feedforward compensator, calculating the current feedforward amount in each control cycle and superimposing it on the current loop command; calculating the friction compensation amount online according to the real-time speed command and injecting it into the servo system, working together with the original PI feedback control to offset friction disturbances; this invention, through a simplified segmented static model and parameter identification based on steady-state response, greatly reduces the complexity of the algorithm and the difficulty of engineering implementation while ensuring the compensation effect.

This invention discloses a method, modeling method, system, and application for predicting dynamic contact angle distribution, belonging to the technical field of predicting dynamic contact angle distribution of micro-protrusions. The method for predicting dynamic contact angle distribution provided by this invention constructs a joint probability density function of horizontal distance and height difference based on the normal distribution law of the horizontal distance and height difference between adjacent micro-protrusions on a rough surface; derives the probability density function of the static micro-protrusion contact angle through integral transformation and Leibniz's differentiation rule; derives the first and second relations through cyclic load experiments; and constructs a probability density function model of the dynamic contact angle based on the first and second relations and the probability density function of the static contact angle. By integrating the static model and the dynamic relations, it achieves the prediction of contact angle distribution under different cycles, adapting to the evolution of contact characteristics under varying loads, and effectively solving the problem that traditional static models cannot reflect the dynamic evolution of the contact angle under cyclic loads.

The application discloses an industrial production adaptive statistical process control method and system based on causal inference and dynamic pattern recognition, and belongs to the technical field of industrial production process quality monitoring. In view of the problems of static control limit, poor working condition adaptability and difficult fault root cause positioning in the prior art, the application integrates data preprocessing, multi-working condition recognition, dynamic feature extraction, adaptive multivariate monitoring, abnormal pattern recognition and early warning, causal inference diagnosis and online learning updating and other methods and function modules. Through deep learning, time sequence features are extracted, a sliding window and a clustering algorithm are used to recognize a running working condition, dynamic control limits are introduced to adapt to process changes, and fault root cause positioning is performed in combination with causal testing and a Bayesian network. The application effectively solves the problem that a static model cannot adapt to working condition migration, and significantly improves the abnormal detection precision, fault diagnosis efficiency and adaptive monitoring capability of a complex industrial production process.

This invention belongs to the field of industrial internet security, specifically involving an industrial internet security capability assessment framework aimed at solving the problems of data fragmentation, static models, lack of cross-node coupling relationships, and difficulty in quantifying resilience. The invention includes: collecting asset fingerprints, controlling traffic, alarms, and operation and maintenance logs; integrating process roles and instruction sequences to construct a baseline of business assets and traffic triples; quantifying compliance and boundary isolation, probe and traffic coverage, and closed-loop processing time for assets as nodes; calculating the resilience coefficient of steady-state recovery time under virtual disturbances based on a collaborative influence matrix and causal graph; aligning by process cycle and dynamically correcting weights using causal discovery; and integrating weights and resilience output for phased assessment and grading. This invention achieves deep fusion of multi-source data, collaborative capability assessment, and resilience quantification, and adaptively adjusts with business changes, improving the accuracy and real-time performance of the assessment.

The present application relates to a kind of source pyrolysis gas and source outside pyrolysis gas quantitative evaluation method, comprising: first, respectively establish the mathematical model describing the change of two types of gas products with geological evolution;Second, based on the results of mathematical model, build the analysis chart of gas composition under different mixing ratio;Finally, the measured gas sample data is projected on the analysis chart, i.e. the proportion of two types of source gas is determined, and the quantitative evaluation of source pyrolysis gas and source outside pyrolysis gas is completed.The present application constructs a dynamic evolution model reflecting geological history to replace the traditional static model, and generates a gas source contribution ratio identification chart based on the model.Through the measured data point projection, quantitative analysis can be completed, which overcomes the limitations of traditional methods and improves the accuracy and reliability of identification.

This invention discloses an optimization method for co-firing coal and biomass gasification gas based on dynamic modeling, including S1, data acquisition and preprocessing; S2, dynamic characteristic modeling; S3, online updating of model parameters; S4, setting optimization objectives and constraints; S5, multi-objective dynamic optimization solution; S6, generation and issuance of optimization instructions; S7, real-time adjustment of actuators; S8, closed-loop feedback monitoring; and S9, periodic verification of the model and strategy. By establishing and updating a dynamic mathematical model online, this invention can accurately capture the dynamic response of the combustion system as fuel characteristics and load change, achieving real-time and forward-looking optimization of the co-firing ratio. Furthermore, this method effectively overcomes the poor adaptability of static models, significantly improving the overall thermal efficiency of the co-firing process while ensuring that pollutant emissions are stably controlled at a low level.

This invention relates to the fields of computer vision and 3D graphics technology, specifically disclosing a target identity and trajectory localization method driven by multi-camera 3D reconstruction. The method involves acquiring video sequences through a synchronous camera network and constructing an implicit representation model containing only static background structures to establish a stable 3D coordinate system. Then, target detection is performed on the real-time video stream, and the 2D coordinates of the detection points are back-projected and the accurate 3D spatial coordinates are calculated by querying the static model. Finally, the target's motion state, appearance, and behavioral characteristics are fused to perform multi-target tracking and identity association, generating a 3D motion trajectory with a consistent identity identifier.

The application provides a user loss early warning method and device, equipment and a storage medium, and relates to the technical field of user operation. The method comprises: obtaining behavior data and device attribute data of a target user at multiple historical times for an Internet TV system; obtaining user time sequence data of the target user according to the behavior data and the device attribute data at the multiple historical times; and predicting, by using a loss early warning model, a loss probability and a loss reason of the target user according to the user time sequence data. The application can accurately depict the gradual process of user behavior from active to recession, solve the problem that a static model cannot capture dynamic changes, realize early identification and accurate attribution of loss risk, and thus intervene in advance.

The application discloses a laboratory multi-source heterogeneous data AI analysis and detection process linkage method based on a knowledge graph, and belongs to the technical field of electronic digital data processing, and comprises the following steps: a four-dimensional knowledge graph and an atomized AI primitive library are constructed, and a computable knowledge-ability mapping is formed. By receiving a high-level detection intention and analyzing the intention into an intention shape, the target is reversely decomposed into sub-targets and semantically matched with AI primitives under the guidance of the graph, a candidate process skeleton is generated through symbolic reasoning, a graph neural network proxy model and Bayesian optimization are used for neural level parameter optimization, and neural-symbolic collaborative program synthesis is realized. Finally, the synthesized process is subjected to multidimensional feasibility verification, and the optimal one is executed, and the successful process is used as new knowledge to feed back to the graph, so that a closed loop is formed. The application breaks through the limitation of a traditional static model library, so that the system can automatically synthesize a new detection process meeting semantic and performance constraints according to dynamic and open field logic.

The present application relates to the technical field of comprehensive energy, and provides a safety evaluation method and system for comprehensive energy system planning based on energy path theory, a continuous flow model is constructed by combining a steady-state alternating current flow model of a power system and a dynamic gas flow model of a natural gas system, the limitations of traditional static models that cannot depict the dynamic differences of multi-energy systems are broken through, the system response characteristics under faults are more truly reflected, the linear sensitivity method of saddle node bifurcation point singularity is used to realize rapid preliminary screening of faults, the load margin is accurately estimated by combining generalized curve fitting to complete secondary screening, and the final fault set is accurately evaluated through dynamic continuous energy flow analysis, which not only solves the problems of existing partial optimism and insufficient dynamic response depiction, but also considers the influence of electric-gas faults in a unified framework, gradually identifies key vulnerable lines and load margin indicators, and significantly improves the accuracy and reliability of comprehensive energy system safety evaluation.

This application provides a network security threat monitoring method, system, device, medium, and product. The method integrates multimodal raw data, performs parsing and refinement, standardizes ontology architecture definition, and aligns sparse anchor points with topology guidance to achieve cross-platform and cross-protocol semantic fusion. Then, it constructs a dynamic spatiotemporal heterogeneous graph based on the fused entity relationship data and performs streaming incremental updates to form a unified security graph. It uses graph neural networks and spatiotemporal association models to detect collaborative attack behaviors. Through adversarial elastic embedding and reinforcement learning mechanisms, it achieves robust source tracing and family merging of attacks with structural disturbances or temporal variations. Finally, it generates structured disposal suggestions based on the metagraph mechanism and displays them visually. This solves the problems of traditional security monitoring systems, such as difficulty in identifying collaborative attacks due to reliance on a single data source, difficulty in modeling complex spatiotemporal dependencies with static models, low detection rate and high false alarm rate for malicious behaviors and evasion methods, lack of elasticity in source tracing and merging, and unstructured disposal suggestions.