395 results about "Reduced model" patented technology

The invention discloses a transmission tower vulnerability analysis method and system based on seismic simulation, and belongs to the technical field of electric power facility disaster prevention, and the method comprises the steps: correcting bedrock seismic oscillation parameters based on equivalent shear wave velocity, and generating a high-precision earth surface PGA distribution diagram; automatically positioning a key node layer through eigenvalue decomposition, reducing the order of the three-dimensional model of the power transmission tower into a two-dimensional series multi-degree-of-freedom model, and outputting a layer mass vector, a condensation stiffness matrix and rod piece bending stiffness; performing time-history analysis by using the corrected PGA-driven reduced-order model to generate an earthquake damage probability; the multi-source remote sensing data and the PGA are fused, the landslide space probability is output through a deep learning model, and the impact strength and the landslide kinetic energy are calculated to judge the landslide damage probability; and based on the dual-threshold condition triggering probability union set, calculating a comprehensive damage probability, and generating a three-dimensional vulnerable curved surface. According to the method, the problems of high missing report rate and low calculation efficiency of single disaster assessment are solved, and minute-level accurate early warning of the composite disaster risk of the power transmission tower is realized.

The invention requests to protect an unmanned aerial vehicle light-weight small target detection method based on a YOLO-GLL neural network, and the method comprises the following main steps: S1, constructing a YOLO-GLL target detection model based on YOLOv8s, and inputting a training set image to complete model training; s2, extracting different scale feature maps of the image through a backbone network containing a partially grouped multi-scale convolution module; s3, performing cross-level fusion on the feature map by using the lightweight multi-scale fusion feature pyramid neck network; s4, executing target classification and bounding box regression by adopting a lightweight shared detail enhancement detection head to obtain a lightweight high-precision model; and S5, inputting a to-be-detected unmanned aerial vehicle image, and realizing small target detection and identification in combination with the Soft-NMS and Shape-IoU post-processing strategies. The method has the advantages that (1) the detection precision is improved; (2) the parameter quantity of the model is reduced, and the deployment of resource-constrained equipment is adapted; and (3) a post-processing strategy is improved, the problem of missing detection of small targets in a dense shielding scene is relieved, and the robustness of a complex scene is enhanced.

The invention discloses a welding simulation heat source parameter calibration method and system based on a hybrid agent model, and belongs to the technical field of welding numerical simulation and optimization, and the method comprises the following steps: S1, problem definition and experiment design; s2, analyzing a product structure and constructing a joint model; s3, parameter space sampling; s4, performing simulation calculation; s5, constructing a mixed agent model; s6, proxy model optimization based on a genetic algorithm; s7, parameter verification and model updating; and S8, optimizing and outputting optimal parameters. According to the method, most time-consuming full-order finite element calculation is replaced by the proxy model and the simplified joint model, the parameter calibration process is shortened from several days or weeks to several hours, and the optimization efficiency is improved; a simplified model and a complete model are used for optimization, firstly, the simplified model is used for rapidly obtaining an optimal parameter approximate value, and the complete model is used for obtaining accurate parameters.

The invention discloses a multi-variable measurement sensor state lightweight evaluation method based on knowledge distillation, and belongs to the technical field of electric digital data processing and multi-sensor data fusion. The method comprises the following steps: firstly, carrying out time synchronization and physical consistency constraint modeling on original data of multiple sensors, and extracting feature representation; a high-precision teacher model is trained at the cloud end, and the high-dimensional mapping relation of the sensor state is learned; intermediate features, soft output and uncertainty information of a teacher model are extracted to serve as distillation knowledge, lightweight student model training is guided, and effective migration of discrimination knowledge is achieved in combination with soft label constraint, feature alignment and an uncertainty guiding mechanism; and finally, compressing, quantifying and optimizing the student model, and deploying the student model to a vehicle-mounted end to realize real-time evaluation and dynamic updating of the states of multiple sensors. According to the method, the model complexity is greatly reduced while the evaluation precision is ensured through a knowledge distillation framework, and efficient and reliable state perception and fault-tolerant control support is provided for an intelligent driving system.

The invention discloses a flywheel system bearing assembly multi-dimensional rule reduction health state testing method, and relates to the technical field of flywheel system bearing assembly health state testing. Acquiring an actual measurement data set of the flywheel system bearing assembly; determining a confidence coefficient corresponding to the reference value set and the result, and constructing an initial health state test model based on a confidence rule base; constructing a multi-dimensional rule evaluation framework; performing composite activation on the reduced model; optimizing parameters in the model by adopting an optimization algorithm; and reasoning the model by using an evidence reasoning analysis algorithm to obtain a health test result of the flywheel system bearing assembly, and realizing an accurate health state test through the evidence reasoning algorithm.

The invention discloses an air inlet casing full-process simulation closed-loop optimization method, which relates to the technical field of mechanical manufacturing and processing, and comprises the following steps: simplifying an air inlet casing model; selecting a plurality of key procedures from the whole procedures; according to process parameters and boundary conditions of the key process, grid division is carried out on the simplified model, and a corresponding finite element simulation model is constructed; sequentially simulating the finite element simulation model according to a key process sequence, correcting a simulation result according to deformation data of actual processing, transmitting the corrected simulation result to a next key process for model simulation until simulation and correction of all key processes are completed, and constructing a full-process simulation prediction model; performing iterative optimization on multiple parameters in the whole-process simulation prediction model to obtain an optimal parameter combination; and correcting the simulation model according to the deviation between the actual processing data and the simulation result under the control of the optimal parameter combination. According to the method, high-precision prediction of various complex manufacturing processes can be realized, and the manufacturing precision is improved.

The invention relates to an air-ground integrated three-dimensional scene automatic modeling method, system and device and a storage medium, and the method comprises the steps: carrying out the point cloud fusion and coordinate registration of a preprocessed air original point cloud and a preprocessed ground original point cloud, and obtaining an air-ground fusion point cloud image; performing semantic segmentation on the air-ground fusion point cloud image through a pre-trained point cloud semantic segmentation network to obtain a composite point cloud cluster containing multiple semantic categories; extracting a building outer contour of the building type point cloud cluster, simplifying the building outer contour according to a preset simplification rule, and constructing a building three-dimensional simplified model according to the simplified outer contour; extracting category feature parameters of the non-building type point cloud clusters, searching a standard model conforming to the category feature parameters from the preset model library, and replacing the corresponding point cloud clusters with the standard model to obtain a non-building three-dimensional simplified model; and performing archiving according to the semantic layer to form a structured three-dimensional model of the to-be-modeled urban area.

The present invention discloses a smart bio-inspired material design platform to satisfy multi-objective material design featuring complex microstructure for the future. the platform sets mechanical properties of a simulative material element via establishing a reduced model. A distribution of the simulative material element is simulated so as to output a material simulative parameter. A deep learning framework is combined in the platform for computing and evaluating an optimal material design that meets a target material parameter. Specifically, the reduced model can be based on data provided by any test of material mechanical properties, and the deep learning framework evaluates whether a biomimetic material design meets demand of the optimal target material parameter according to a standardized reward function model. The platform is applicable to multi-objective simulative material design, and is greatly potential for futuristic applications.

The invention relates to the field of ecological environment and digital simulation, in particular to a phosphorus pollution risk evaluation method and system based on a migration and transformation mechanism. A phosphorus pollution risk evaluation method based on a migration and transformation mechanism comprises the following steps that S10, at a preset sampling point position of a preset drainage basin, various data are collected according to preset parameter types, and the parameter types comprise various parameter types needed by a preset mechanism model and a preset output coefficient model; sorting the collected data according to a collection time sequence, and establishing a space-time matching database; and S20: extracting parameter types in the database, defining a cross-medium migration relationship of phosphorus based on physical parameters, constructing a conversion relationship of a corresponding form in combination with parameters of chemical and biological types, and dividing factor types. According to the scheme, the problem of low evaluation accuracy caused by insufficient sensitivity of analysis results of a static evaluation method and a dynamic evaluation method based on a simplified model in phosphorus pollution risk evaluation is solved.

The invention discloses a dense topology model lightweight method, and belongs to the technical field of three-dimensional model processing. The problems that an existing scheme is low in efficiency, large in structural damage, poor in mapping quality and the like are solved. The method comprises the following steps: importing a dense grid model containing high-precision geometry and texture; selecting an automatic (adaptive lattice clustering algorithm) or manual mode to generate a target simplified topology; constructing a simplified model fitting the original model through a multi-direction projection technology; uV expansion is realized through automatic blocking, LSCM algorithm optimization and manual adjustment, and textures are baked in combination with RendertoTexture and SSAA technologies; and after performance optimization, outputting a lightweight model compatible with a mainstream format and an engine. According to the invention, high efficiency, stability and light weight are realized, key features are accurately reserved, visual consistency is guaranteed, and multi-scene application is adapted.

The invention discloses a large model parameter protection method based on singular value decomposition in a trusted execution environment. Positioning a key layer of the large model through gradient sensitivity and interlayer correlation measurement in a large model fine tuning training process, and decomposing a weight matrix of the key layer by using singular value decomposition to obtain sensitive parameters and non-sensitive parameters; dividing a large model operation environment into a rich execution environment and a trusted execution environment, and generating a reversible key matrix in the trusted execution environment; when input data pass through the key layer for the first time, the rich execution environment uses non-sensitive parameter reasoning and then transmits the input data to the trusted execution environment for sensitive parameter reasoning, then the data are encrypted by using the key matrix, the encrypted data are transmitted back to the rich execution environment for reasoning, and when the input data pass through the key layer again, the rich execution environment conducts reasoning and then transmits the input data to the trusted execution environment. And the final data is output after decryption and reasoning. According to the method, the side channel attack risk is reduced, the data transmission security and the calculation efficiency are improved, and the model reasoning delay is reduced.

The invention discloses a hierarchical simulation modeling system for a new energy station cluster and a real-time verification method, belongs to the technical field of new energy power generation simulation, and solves the problems of high simulation calculation complexity, difficulty in real-time operation and closed-loop interaction verification with an external control system of a large-scale new energy station cluster. According to the technical scheme, the method comprises the steps that a digital dynamic real-time simulation platform configured with a multi-thread parallel computing architecture is adopted, and a new energy station detailed model comprising a wind power generation unit model, a photovoltaic power generation unit model, an energy storage power generation unit model and a current collection system model is deployed on the platform; generating an independent wind speed sequence for each fan through a wind resource distribution model; and constructing a large-base station cluster model composed of a plurality of station simplified models through aggregation equivalence. The system is mainly used for real-time simulation testing of a large-scale new energy base and closed-loop verification of a control system.

The invention relates to the technical field of building information model optimization, in particular to a large-volume BIM model optimization system based on intelligent loading, which comprises a model analysis unit, a function identification unit, a simplified execution unit and a model recombination unit, and is characterized in that the model analysis unit reads an IFC file and separates component data; the function identification unit outputs a decoration component (first class) set, a non-bearing structure (second class) set and a bearing component (third class) set through geometric and semantic dual-channel analysis, the simplified execution unit performs edge folding on the first class, performs vertex clustering on the second class and skips simplification on the third class, and the model recombination unit binds metadata and constructs a dynamic octree index. The problems that a traditional simplification strategy is single and core information is lost are solved, the model loading efficiency and application precision are improved, and the method is suitable for complex building scenes.

The invention relates to the field of natural language processing, in particular to a low-rank adaptation system capable of differential polarization rank search, which comprises an initialization module, a warm-up training module, a joint optimization module, a sparse pruning module and a final fine tuning module. According to the method, dynamic rank allocation and continuous search are realized by introducing trainable rank architecture parameters, and model parameter waste and calculation overhead are remarkably reduced by combining polarization regularization and L2 regularization strategies. According to the method, the performance of the large model in the downstream task can be effectively improved, meanwhile, the method is suitable for scenes with multi-task concurrence, resource-constrained device deployment and high real-time requirements, and a technical solution is provided for efficient adaptation of the large-scale pre-training model.

The invention relates to the technical field of integrated energy systems, and particularly provides a multi-energy-flow scheduling method and system for a regional integrated energy system, and the method comprises the steps: S1, carrying out the unified modeling of a DC micro-grid, a heat supply network and a natural gas network based on a circuit comparison method, and obtaining a network model of each energy network; s2, constructing a direct-current micro-grid island hierarchical control strategy; and S3, according to the state parameters of the regional integrated energy system, performing dynamic load flow calculation on the direct-current microgrid under the island hierarchical control strategy based on a decomposition load flow method and the network model of each energy network, performing steady-state load flow calculation on the heat supply network and the natural gas network, and generating a scheduling instruction, so that multi-energy-flow coupling response can be accurately captured, and the scheduling efficiency is improved. According to the method, the model inconsistency error is reduced, the unit calculation burden is reduced, the expandability and fault tolerance of the system are improved, and high-precision support is provided for optimal scheduling and long-term planning.

Systems and methods provide a HybridOps model for the identification, capture, isolation, feature engineering and adjudication of source signal data signatures for inclusion in calibration quality standard reference signal data signature libraries that improve machine learning and validation, reduces model bias and reduces model drift. The HybridOps model may include an “unlocked” AI / ML (machine learning enabled) public facing deployment pipeline in parallel with a clone AI / ML deployed in an internal development environment using a ML-Ops pipeline and in parallel with a clone“locked” AI / ML (machine learning disabled) as a standard reference. The three deployed models enables monitoring and measuring model drift, context drift and product progression for improved verification and validation of model reliability. The parallel environment AI / ML testing using randomized learning, validation, testing sets drawn from calibration quality adjudicated standard reference signal data signature libraries provides for validation, verification and test to failure procedures.

The present invention relates to the technical field of building information modeling (BIM). Disclosed is a BIM lightweight method based on fine-grained geometric objectification, the method comprising the following steps: S1, creating a mapping table T to store data of Face objects and Mesh objects in a model; S2, calling an IExportContext interface in a Revit SDK to extract data of components in the model; S3, checking whether the processing of the components in the model is currently completed; S4, if the processing is completed, ending the process; and if the processing is not completed, acquiring the value of a current flag; S5, if flag=1, calling a Face object lightweight processing process and returning to S2; and S6, if flag=0, calling a Mesh object lightweight processing process and returning to S2. The present invention can effectively simplify model data and realize model lightweighting; moreover, obtained data supports LOD technology, which can greatly reduce resource consumption during front-end rendering and improve rendering efficiency.

The invention discloses an AI model parameter initialization method based on model parameter and structure multi-modal fusion, and belongs to the technical field of artificial intelligence. The method comprises the following steps: firstly, collecting historical pre-training model data of a cross-model architecture and a cross-data set, processing model parameters into a token sequence, training a Transform codec, converting a model structure into a graph, and training GAT to extract structural features; a multi-modal feature data set is constructed after a related network is frozen, structural features serve as core conditions, a conditional diffusion model DDPM is trained through AdaLN modulation and residual module coupling, and multi-modal feature fusion of'parameter feature-structural feature 'is established. In the reasoning stage, the structural features of the unknown model are extracted, the random parameters of the unknown model are combined with the submerged space shape determined by the encoder, sampling is conducted through the conditional diffusion model, anti-token processing is conducted through the decoder, and adaptive initialized model parameters are generated for the unknown model. According to the method, cross-model structure high-quality parameter initialization is realized, the model training time is shortened, and the large model training requirement is met.

The invention discloses a crystal grain detection and analysis method and related equipment, effective crystal grains are obtained through corrosion expansion pretreatment denoising, OfficientNet and bidirectional feature pyramid network detection, contour obtaining through region segmentation and three-step filtering, qualification is judged in combination with a threshold value, a report is generated, and the existing technical problems are solved in a targeted mode. On the aspect of reducing calculation complexity, preprocessing reduces model data volume and interference, the model abandons complex modules, parameters are few, calculation cost is low, hardware occupation can be reduced, and detection and training time can be shortened; on the aspect of improving generalization ability, features are preprocessed and purified, interference of imaging conditions and material types is reduced, the number of layers of the model is simplified, multi-scale fusion ability is achieved, and detection in different fields can be adapted without optimization of a specific data set; on the aspect of reducing data dependence, invalid and pseudo crystal grains are accurately removed through three-step filtering, defects of a small-scale or low-quality data set are cooperatively made up, dependence on a high-quality large-scale data set is greatly reduced, and robustness is improved.

The invention discloses a three-dimensional model basic body simplification method and system, and relates to the technical field of three-dimensional model simplification. According to the method, an improved PointNet + + semantic segmentation network is constructed, an attention mechanism is introduced, and normal vector and curvature input is added, so that refined semantic segmentation and component marking of an original three-dimensional model are realized; extracting a core body based on a semantic tag and a non-body structure exclusion rule, and calculating a bounding box of the core body; performing dynamic precision grid division and performing interference check in combination with a semantic rule to generate a three-dimensional placeholder matrix; performing fitting and Boolean operation on the occupied area by using the parameterized basic body to obtain a simplified model main body; assembling and combining the key feature part and the simplified main body, and outputting a final simplified model; according to the method, the defects of blind simplification and low fidelity in the prior art are overcome, automatic and high-fidelity model simplification is realized, and the efficiency of virtual assembly verification and the model reusability are remarkably improved.

A system and method for neural network confidence regularization is disclosed. A classification system uses a neural network training model to generate prediction. The confidence of the neural network training model is adjusted based on feature prevalence. The system processes mixed data types (binary, categorical, continuous, and date) through type-specific transformations and tensor construction. A tempering factor is calculated from the unweighted sum of features and applied to intermediate neural network outputs. This tempering mechanism reduces model confidence when several low-weight features are present, enabling faster convergence, better generalization, and improved classification accuracy compared to standard neural networks, particularly for complex non-linear relationships in tabular data domains.

A machine learning legal natural language generation computing system for developing a pre-trained and instruction fine-tuned MLLNLG generative language model that aligns with human instructions and preferences using causal language modeling for next-token prediction. The training method employs a scaled version of RoPE to compress token position indices, to manage longer sequences beyond the computing hardware devices (GPU's) DRAM memory capacity. This system is optimized for various legal NLP and NLG tasks by receiving input tasks from users through an I / O component. The system enables efficient pre-training of a generative legal language model from scratch on a single computing hardware device (GPU), achieving an efficient MFU of 41.35 and through weight tying, the system reduces model parameters compared to LLMs, eliminating the need for multiple GPUs. Tailored for the legal domain, the model excels in processing legal jargon and demonstrates advanced reasoning capabilities, outperforming existing models in summarization tasks.

The invention relates to the technical field of sewage treatment, and discloses a sewage system deposition problem evaluation method and device, a medium and a program product, and the method comprises the steps: constructing a target sewage system model capable of coupling a sediment movement process through integrating a sewage discharge rule, a real-time multi-source data set, a historical data set and a preset organic matter proportion parameter; the method breaks through the limitation that a traditional evaluation method only depends on single data or a simplified model, and achieves the comprehensive, dynamic and accurate evaluation of the deposition problem of the sewage system. Historical data and real-time data are combined, so that the model has historical experience support and can reflect the current system state, the introduction of preset organic matter proportion parameters considers the influence of organic matters in sewage on deposition, the model is more suitable for the actual situation, the reliability and accuracy of an evaluation result are greatly improved, and the method is suitable for popularization and application. And a scientific basis is provided for prevention and treatment of the deposition problem of a sewage system.

The invention relates to the technical field of thermal management, in particular to a liquid cooling plate flow resistance characteristic evaluation method and device, a medium and equipment. According to the method, the calculation load is remarkably reduced by converting the flow data into the reduced-order model by utilizing the intrinsic orthogonal decomposition. This can achieve faster analysis without sacrificing accuracy, so that the design can be evaluated and optimized in real time. Moreover, different from the traditional method which may depend on a simplified model or empirical correlation, the method based on the intrinsic orthogonal decomposition can capture a main flow mode, namely a dominant mode, which causes resistance. These modes reveal critical flow characteristics, such as high shear stress, vortex formation and flow separation, which are crucial for understanding the root cause of flow resistance. The capability of quantifying and isolating the important flow structure can deeper understand how the specific flow behavior affects the resistance, so that guidance is provided for the optimization design of the geometrical shape of the liquid cooling plate.

The invention relates to a multi-scale feature recognition method for power equipment in the technical field of intelligent detection of the power equipment. Aiming at the problems of high calculation complexity, insufficient real-time performance and weak multi-scale target identification capability of a traditional electric power detection model, the method is optimized through the following technical scheme: replacing a self-attention mechanism in a Grouping-DINO model with a Comba recursive structure, constructing a lightweight student model, and constructing a multi-scale target identification model; knowledge migration from a teacher model to a student model is realized through a multi-task distillation loss function, a four-scale feature pyramid network is added behind a backbone network of the student model, and feature extraction and key area focusing of a multi-scale target of power equipment are enhanced in combination with a channel self-attention weighting mechanism. The method effectively maintains the cross-modal feature fusion capability and detection precision of the model while remarkably improving the reasoning speed and reducing the model volume, and is particularly suitable for real-time detection of power equipment in edge calculation scenes such as unmanned aerial vehicle inspection.

The invention discloses a heterogeneous fuzzy and game-based double-layer adaptive evolution CAD modeling command generation method, and belongs to the technical field of artificial intelligence and computer-aided design intelligent modeling. The method solves the technical problems that in an existing natural language driven CAD command generation technology, fuzzy semantics are difficult to quantify and map, heterogeneous commands are poor in adaptability, the executable performance of generated commands is low, and the reasoning process is black-box-shaped. According to the method, a semantic analysis-game generation-evolution refinement-verification backtracking-knowledge precipitation full-process closed-loop architecture is constructed, the method can be applied to the fields of industrial software, three-dimensional modeling, intelligent design systems and the like, the engineering performability, generalization ability and reasoning interpretability of command generation are improved, the generation diversity and feasibility are balanced, and the method is suitable for popularization and application. And the model iteration cost is reduced, and technical support is provided for industrial landing of the text-driven CAD modeling technology.

The invention discloses a global reduced-order perception self-adaptive precise temperature control method. The method comprises the following steps: establishing a heat transfer mathematical model of a precise temperature control object and a thermal environment boundary thereof; performing spatial discretization processing on the heat transfer mathematical model to obtain a heat balance equation; constructing a discrete state space model based on a heat balance equation; carrying out order reduction processing on the discrete state space model based on a Pitot number criterion; arranging a temperature sensor and a heating loop based on the model after order reduction; carrying out dimension expansion processing on the model after order reduction; performing real-time filtering and parameter estimation on the extended state space model by using an extended Kalman filter to obtain posterior estimation values of the temperature and the heat exchange coefficient; and calculating a control quantity according to the posterior estimation value to realize feedforward-feedback composite control. According to the temperature control method provided by the invention, muK-level, high-dynamic and self-adaptive temperature control on the spacecraft precision load instrument under the conditions of sparse sensing and strong noise can be realized.

The invention discloses a three-phase inverter few-sensor model prediction control method and device. The method comprises the following steps: acquiring voltage and current sample data based on a three-phase inverter model prediction control system; aiming at the sample data, constructing and training a neural network; and voltage and current signals are calculated by adopting a neural network, and model prediction control is implemented. According to the designed method, the number of sensors of the three-phase inverter model predictive control system is reduced, and the anti-interference performance of the system is improved.

The invention discloses a dynamic temperature regulation and control method for a mass concrete cooling water pipe of an anti-collision pier bearing platform. The method comprises the steps of deployment of a monitoring module, preprocessing of multi-source monitoring data, deployment of an execution module, establishment of a neural network model, training of the neural network model, real-time continuous decision making and dynamic temperature regulation and control. A neural network model based on physical constraints ensures that the regulation logic conforms to the thermodynamic law; through historical multi-working-condition data training, the model can respond to the environment change and the hydration process in real time, the optimal regulation and control parameters are output, the limitation of a traditional empirical formula or a simplified model is solved, and dynamic optimization regulation and control are achieved.

The invention discloses a liquid cooling heat dissipation assembly brazing filler metal filling analysis method based on multi-physics field coupling simulation, and the method comprises the steps: obtaining a three-dimensional CAD model of a liquid cooling heat dissipation assembly, and simplifying the three-dimensional CAD model; establishing a physical field coupling simulation model, setting simulation initial conditions and boundary conditions for simulating brazing filler metal flow, and setting simulated brazing process parameters; running the simulation model in Fluent software to obtain a flow path parameter, a filling speed parameter, a brazing filler metal temperature cloud picture and a second-phase brazing filler metal volume fraction cloud picture of the brazing filler metal in the simplified model, judging whether a simulation result is incomplete filling or blockage or not, performing verification through actual operation, and confirming the consistency of the simulation result and actual verification; and optimizing the process parameters, and outputting a simulation result as the completely filled process parameters. According to the method, the simulation model is established, brazing filler metal filling analysis can be achieved, verification of actual experiments is conducted, the accuracy of the model is guaranteed, then technological parameters are optimized, and brazing filling is complete.