3760 results about "Training data sets" patented technology

A method, system, and device for wind speed prediction and layout optimization in wind power generation are provided. The method includes: obtaining a basic wind resource dataset of a target region; constructing a physics-informed neural network model based on the basic wind resource dataset; obtaining wind speeds data at a specific location in a velocity field based on the physics-informed neural networks and constructing a training dataset; training the physics-informed neural network model based on the training dataset; reconstructing a wind speed distribution within the velocity field and predicting wind speeds for a next time period with a wind farm using the trained physics-informed neural network model; and optimizing a layout of a wind turbine cluster based on a reconstructed wind speed distribution within the velocity field. The present application reconstructs a two-dimensional velocity field of the wind farm by training the PINN and enables accurate ultra-short-term wind speed prediction.

The invention discloses a heterogeneous road structure cavity disease inspection method, device and equipment and a storage medium, and the method comprises the steps: constructing a highly-simulated heterogeneous road structure model, and adaptively generating an irregular cavity disease meeting the horizon geometric constraint; the problems that electromagnetic wave reflection simulation is distorted due to the fact that a traditional homogeneous simplified model cannot reflect dielectric mutation between aggregate and a cementing material, and a regularized cavity form does not conform to a real stress distribution expansion mode are solved. According to the method, radar forward modeling simulation data containing complex heterogeneous backgrounds and randomly-shaped holes can be automatically generated in batches, a rich, real and diversified training data set is provided for intelligent identification of road internal diseases based on deep learning, and the bottleneck problem of lack of real disease radar image data is relieved.

The invention discloses an industrial Internet of Things time sequence self-supervision anomaly detection method and a monitoring and early warning system, and relates to the field of industrial Internet of Things, and the method comprises the steps: S1, constructing an anomaly detection model, and S2, obtaining a training data set; s3, training and optimizing an anomaly detection model; s4, acquiring to-be-detected data in real time; s5, performing anomaly detection analysis on the to-be-detected data, and outputting an anomaly detection result; through a time sequence and relation learning module, a dynamic graph topological structure learning module and an enhancement module, internal characteristics of a time sequence in a time domain and a space domain are deeply mined. The time sequence and relation learning module comprehensively captures a multi-scale time pattern, and the dynamic graph topological structure learning module eliminates dependence on a predefined graph structure; the enhancement module enhances the invariant representation under noise, and improves the recognition capability of the model to a normal mode; through wide experiments, the advancement of the method in detection performance is verified, and reliable support is provided for intelligent manufacturing and infrastructure diagnosis.

The invention relates to the technical field of coast engineering, in particular to an estuary sandy coast erosion and deposition simulation system based on the coupling effect of flood peak runoff, waves and tide, which comprises a training data generation module, an intelligent agent model construction module, a dynamic prediction engine module and a scene evaluation module. The training data generation module is used for outputting a hydrodynamic state field, a wave characteristic field, a bed surface shear stress field and a bed surface elevation variable quantity by utilizing a traditional numerical model to construct a training data set; the intelligent agent model building module is based on a deep learning network and introduces physical constraint loss function training to obtain an intelligent agent model; the dynamic prediction engine module loads the trained deep learning network to realize bed surface elevation variation prediction and update terrain boundary conditions; and the scene evaluation module executes erosion and deposition evolution simulation according to different hydrological boundary condition combinations and extracts terrain evolution data to quantitatively evaluate the coast stability and the channel deposition risk. The invention relates to the field of estuary dynamic landform simulation and coast engineering.

Various aspects of the present disclosure relate to a use equipment (UE) configured with multiple artificial intelligence (AI) models each of which has been configured (e.g., trained) based on training data sets corresponding to one or more of different conditions, such as location of the UE, orientation of the UE, whether the UE is indoors or outdoors, whether the UE is line-of-sight or non-line-of-sight with a base station, and so forth. A network entity (e.g., a gNB) configures the UE with a set of reference signals for measurement of at least one quantity. The UE generates the measurement report based at least in part on the set of reference signals and one of the multiple AI models, and transmits the measurement report to the network entity (e.g., gNB).

A machine learning method where, in a first step, a first (general) machine learning model is trained using a first training dataset including unlabelled optical fibre sensing data. Then, in a second step, a transfer learning process is applied to adapt or fine-tune the first machine learning model to a more specific application (e.g. to perform a specific type of detection or classification). Due to the large volumes of optical fibre sensing data available, the first machine learning model may provide a general machine learning model which has a high level of generality and is highly adaptable.

The invention provides a digital twin three-dimensional model construction method based on a generative large model, and the method comprises the steps: obtaining multi-source monitoring data of a power distribution network, and processing the multi-source monitoring data into a training data set; the method comprises the following steps: mapping multi-source monitoring data into a multi-scale tensor subspace through tensor wavelet structured transformation, adaptively extracting spatial features through a learnable wavelet kernel, and keeping the structural continuity of a physical field in combination with a geometric prior regular term; constructing and training a generative adversarial network through a training data set; inputting and analyzing the physical parameter vector of the target scene, and if the topological similarity score is lower than a preset threshold value, adjusting noise vector regeneration; and if yes, outputting a three-dimensional model tensor and importing the three-dimensional model tensor into a digital twin platform, and driving real-time physical field visualization. According to the method, characteristics of a multi-scale space structure and a nonlinear physical field can be reserved, the physical rationality and generalization ability of the generated model are remarkably improved, and depth identification of topological attributes (such as hole connectivity and surface defects) and local geometric defects of the three-dimensional model is realized.

The invention belongs to the technical field of artificial intelligence, and discloses a knowledge extraction method and system based on semantic consistency evaluation and hybrid verifiable reward, and the method comprises the steps: constructing a training data set with evidence labeling; based on the training data set, reinforcement learning training is carried out on a pre-trained large language model, a group strategy optimization GRPO algorithm is adopted, and model output is evaluated by using a mixed reward function; and on the basis of an evaluation result of the mixed reward function, updating parameters of a large language model so as to generate structured knowledge which is correct in format, accurate in content and provided with verifiable evidence. According to the method, in reinforcement learning training, effective decoupling format, content and credibility evaluation is realized, and refined feedback is provided for the model; when the model outputs knowledge, traceable original text evidence is provided for the model, so that the credibility and the interpretability of the model are enhanced, the accuracy of outputting the JSON format by the model is improved, and the accuracy and the integrity of the model in the aspect of content extraction are enhanced.

The invention provides a method for identifying PFAS in an environment based on machine learning pseudo-targeted screening, which comprises the following steps: calling mass spectrum data containing PFAS compounds from a preset mass spectrum database, and performing interference peak elimination processing on the mass spectrum data to obtain a model training data set; extracting a feature data set for model training from the model training data set based on a feature extraction standard; training a plurality of machine learning classification models based on the feature data set, and performing performance evaluation on each machine learning classification model based on a training result to determine an optimal machine learning classification model; and analyzing the optimal machine learning classification model, determining key features when the PFAS is screened and identified, and carrying out PFAS screening identification verification on the optimal machine learning classification model according to the key features based on an actual environment sample. The method has the advantages of saving analysis cost, improving analysis efficiency and improving compound recognition accuracy.

The invention discloses a flow field video generation method based on a policy value architecture and online physical exploration, and belongs to the technical field of crossing of artificial intelligence and computational fluid dynamics (CFD), and the method comprises the following steps: step 1, constructing an unsteady flow field multi-modal training data set, step 2, constructing a generative network system based on an Actor-Critic architecture, step 3, constructing an unsteady flow field multi-modal training data set, and step 4, constructing an unsteady flow field multi-modal training data set. Step 4, supervised fine tuning training is carried out in the first stage; step 5, online physical exploration of a generator is carried out in the second stage; step 6, feedback co-evolution of a physical encoder is carried out in the third stage; and step 7, reasoning generation of an unsteady flow field video is carried out. According to the method, a reinforcement learning architecture containing an Actor and a Critic is constructed, a physical equation is packaged into a digital environment, and a training strategy of basic supervision fine tuning, online physical exploration of a generator and coevolution feedback of an encoder is adopted.

The invention discloses an equipment anomaly tracing method and system based on a digital twinborn and graph neural network, and belongs to the technical field of industrial intelligent operation and maintenance and fault diagnosis. The invention provides an innovative solution integrating digital twin high-fidelity simulation and a graph structure deep learning algorithm, aiming at the technical bottlenecks that the generalization performance of an existing data driving method is insufficient under the conditions of fault sample scarcity and category imbalance and the traceability accuracy of unknown and composite faults is poor. The method comprises the following steps: constructing a high-fidelity digital twin integrating multi-dimensional physical attributes and a system topology structure; based on a fault mode, influence and harmfulness analysis method system, constructing a fault mode library comprising a plurality of single fault modes and composite fault modes, and generating an enhanced training data set with accurate labels through an automatic fault injection mechanism; training a graph neural network model with a multi-level attention mechanism by using the data set so as to learn a propagation rule of a fault in a complex system topology; and finally deploying the model to carry out abnormity traceability analysis on real-time industrial Internet of Things monitoring data. According to the method, the fault diagnosis generalization ability and the positioning precision under the sample imbalance condition are remarkably improved.

A driving range simulation method and apparatus for a new energy vehicle, and a medium and a device. The method comprises: splitting a vehicle thermal management system model into four sub-models (110); establishing a training data set for each sub-model (120); establishing reduced-order sub-models for each sub-model (130); and integrating the model, and performing high- and low-temperature driving range simulation on a new energy vehicle (140).

The invention discloses an information extraction data annotation and training system based on a large language model, and relates to the technical field of information extraction, and the system comprises a data set construction module which is used for constructing a pre-training data set and a fine tuning data set; the model continuous pre-training module is used for carrying out continuous pre-training on a preset general large language model based on the pre-training data set to generate a field adaptive pre-training model; the model fine tuning module is used for performing supervised fine tuning training on the domain adaptive pre-training model through a two-stage course learning strategy based on the fine tuning data set, and generating an information extraction model; the retrieval enhancement generation module is used for performing entity-semantic retrieval on an input text based on a preset knowledge base, outputting context information related to the input text, and outputting structured information of the input text based on the context information and an information extraction model, the problems of insufficient generalization ability, poor field adaptability and disastrous forgetting of a general large language model are solved, and the accuracy and robustness of information extraction are improved.

The invention provides a multi-task identification method for strawberry diseases and insect pests, and belongs to the technical field of strawberry leaf disease identification. The method comprises the following steps: marking an enhanced original strawberry leaf image, and constructing a multi-task training data set; inserting a double attention unit on a semantic segmentation model decoder as a segmentation network, constructing a detection network based on a lightweight detection model, inputting a segmentation mask output by the segmentation network into the detection network to construct a cascade model, and performing preliminary training; performing multi-task joint learning on the cascade model; based on the multi-task data set, a progressive training strategy is combined with a cosine annealing learning rate adjustment strategy, and final training is carried out on the cascade model after joint learning; and inputting a to-be-detected strawberry leaf image into the finally trained cascade model, and outputting a disease type and a severity level. According to the method, the recognition sensitivity of tiny disease spots can be effectively improved, and the generalization ability of the model to complex illumination and shielding scenes is enhanced.

The invention discloses a wing design optimization method based on agent-assisted multi-initial-point simulated annealing, and belongs to the technical field of optimization design. Comprising the steps of 1, initializing algorithm parameters and a training data set; 2, constructing an agent model; 3, executing multi-initial-point parallel simulated annealing, and generating a batch of candidate new solution sets; 4, executing a double-elite active learning strategy based on the new solution set, screening the most potential sample to carry out real evaluation, and updating the agent model; 5, cooling the temperature and reducing the step length; 6, judging whether the cumulative evaluation times of the expensive objective function reach the set maximum evaluation times or not; if not, returning to the step 2; if yes, optimization is stopped; and finally, traversing the training data set, and selecting a sample point with the minimum real objective function value as a global optimal solution. According to the method, a multi-initial-point parallel simulated annealing search mechanism and a double-elite active learning strategy are combined, and the global optimal solution is quickly approached under the limited simulation times.

The invention discloses a corroded oil and gas pipeline residual strength prediction method considering a physical constraint loss function, and the method comprises the steps: collecting multi-source feature data of a corroded oil and gas pipeline, obtaining a residual strength measured value as a label, and constructing a training data set; an XGBoost regression model is combined with an SHAP interpretability analysis technology, and the influence degree and the influence direction of each feature on the residual intensity are quantified; constructing a neural network model, and determining an optimal architecture of a neural network by adopting a hyper-parameter optimization method; constructing a physical constraint term based on the influence degree and the influence direction of each feature, introducing the physical constraint term into a loss function of a neural network model, and forming a comprehensive loss function together with a data-driven loss term; and training the optimized neural network model by using a comprehensive loss function to obtain a final residual intensity prediction model. The method has the advantages that the prediction precision is improved, the model interpretability is enhanced, overfitting is prevented, and multi-source feature data are effectively integrated.

The invention discloses a sewage high-risk pollutant screening and identification method based on fragmentation tree pre-training, and the method comprises the steps: constructing a fragmentation tree pre-training data set according to the second-level mass spectrum data of known high-risk pollutants, so as to carry out the self-supervision pre-training of a graph neural network encoder, and obtaining a fragmentation tree editor; the method comprises the following steps: acquiring secondary mass spectrum data of suspected high-risk pollutant related compounds in a to-be-detected sewage sample, constructing a fragmentation tree set, constructing a high-risk pollutant screening model based on a fragmentation tree encoder, performing high-risk pollutant signal screening on the fragmentation tree set to obtain a high-risk candidate fragmentation tree set, and performing high-risk pollutant signal screening on the high-risk candidate fragmentation tree set. And generating a candidate molecule set of each high-risk candidate fragmentation tree, and identifying a target molecular structure and a matching score of the high-risk candidate fragmentation tree from the candidate molecule set. According to the invention, rapid screening and priority ranking of high-risk pollutants in sewage are realized, and candidate output of structure identification is further provided on the basis of screening.

The invention discloses an unmanned aerial vehicle navigation method and related equipment based on a visual language model, and the method comprises the steps: carrying out the thinking chain construction of an initial training data set, generating high-quality thinking chain data, and constructing a target training data set according to the initial training data set and the high-quality thinking chain data; based on a supervised fine tuning method and a reinforcement learning method, training the initial visual language model according to the target training data set to obtain a candidate visual language model; deploying the target visual language model passing the model verification to an unmanned aerial vehicle navigation control system; and generating a reasoning result and a target position according to the natural language instruction of the user through the target visual language model, and predicting a target action sequence according to the target position by adopting a foresight mechanism, so that the unmanned aerial vehicle executes the target action sequence. Complex natural language instructions can be accurately understood, spatial reasoning is carried out in combination with environmental semantics, accurate target positioning and path planning are achieved, and the method can be widely applied to the technical field of unmanned aerial vehicle control.

The invention discloses a reinforcement learning-based low-voltage power distribution network multi-target collaborative optimization method and system, and the method comprises the steps: S01, taking adjustable equipment of a low-voltage power distribution network as an intelligent agent, and obtaining the historical operation data of a plurality of intelligent agents to form a training data set; step S02, constructing a multi-objective optimization model, and defining a reward function in a phasor form and a multi-objective weight; step S03, performing centralized training on the deep reinforcement learning model, inputting a multi-target weight into a neural network in the training process, performing learning with a maximum accumulated reward according to a reward phasor, and fusing a multi-target weight vector and a Q value vector through a phasor fusion layer; and step S04, acquiring real-time operation data of a plurality of agents in the low-voltage power distribution network, and controlling each agent to act by using a strategy learned by training. According to the invention, multi-target collaborative optimization control can be realized in a distributed photovoltaic large-scale access scene.

A method, computer system, and a computer program product for data-free knowledge amalgamation are provided. Multiple pre-trained teacher machine learning models are obtained. Each is trained on a respective different set of training data. Pseudo-data samples that mimic original training data of the teacher models are generated. A block-wise amalgamation with a self-regulative strategy to integrate knowledge from the multiple teacher models is implemented by inputting the pseudo-data samples into the teacher models and into a student machine learning model. The implementing also includes aligning intermediate representations of the student model with a unified representation capturing relevant features from the teacher models.

The invention discloses a battery capacity loss real-time prediction method based on Arrhenius-LSTM, and the method comprises the steps: S1, collecting the current, voltage, state of charge and temperature data of a vehicle-mounted battery management system, and generating an original time series data set; s2, performing time alignment and capacity calculation on the original time sequence data set to generate a capacity loss sequence; s3, an Arrhenius Arrhenius model is constructed based on the capacity loss sequence, and a physical baseline prediction sequence is generated through temperature related parameter estimation; s4, calculating a difference value between the capacity loss sequence and the physical baseline prediction sequence, and generating a residual sequence; s5, performing feature extraction and time sequence window construction on the residual error sequence to generate an LSTM training data set; s6, inputting the LSTM training data set into the long short-term memory network for training, and generating a residual prediction model; and S7, adding the physical baseline prediction sequence and the output of the residual prediction model to generate a capacity loss prediction result. And the online monitoring and early warning functions of the health state of the battery are realized.

The invention discloses a shale gas well dessert prediction method and device based on integrated learning. The method comprises the steps of obtaining main control factors and feature data of a shale gas well to be predicted, and inputting the main control factors and the feature data into a trained dessert prediction model to obtain dessert prediction data; the model training process comprises the following steps: determining main control factors influencing the dessert according to the correlation between each parameter in the prediction data of the sampled shale gas well in the target area and the dessert; the prediction data comprises first data and second data, the first data comprises geological data, perforation data and oil and gas production data, and the second data comprises logging data and fracturing construction data; performing feature extraction on each parameter in the second data to obtain feature data; constructing a training data set according to the main control factors and the characteristic data of the shale gas well; and training the dessert prediction model based on the training data set. The accuracy of a prediction result can be improved, and fracturing design is effectively guided.

The invention provides a bearing fault diagnosis method based on dynamic hypergraph convolution and spatial-temporal feature fusion. The method comprises the following steps: acquiring a training data set; the training data set comprises a vibration signal and a fault type; constructing a bearing fault diagnosis model based on dynamic hypergraph convolution and spatial-temporal feature fusion; the bearing fault diagnosis model comprises a multi-scale feature fusion module, a dynamic hypergraph learning model, a spatial-temporal feature fusion module and a full-connection classification layer; training the bearing fault diagnosis model based on the training data set; and inputting a to-be-diagnosed vibration signal into the trained bearing fault diagnosis model based on dynamic hypergraph convolution and spatial-temporal feature fusion to obtain a bearing fault type. According to the bearing fault diagnosis method based on dynamic hypergraph convolution and spatial-temporal feature fusion, the problem that the bearing fault diagnosis accuracy is low due to the fact that an existing bearing fault diagnosis method is insufficient in the aspects of multi-damage-degree distinguishing, dynamic feature correlation modeling and physical rule fusion is solved.

Methods for training a series of neural networks to output driving behavior control parameters is disclosed. The training dataset for the neural networks includes sensor-based vehicle driving recordings that may be categorized by geographical area, by qualitative driving behaviors, or by some combination, such that various training data subsets are used to train the series of neural networks. By learning either city-specific driving behavior control parameters, qualitative driving behavior specific driving behavior control parameters, or both, the resulting parameters may then be provided to a motion planning model for use in modeling predictive control for an autonomous vehicle. Rather than relying on XYZ trajectories of agent vehicles when planning future trajectories of the ego vehicle, the motion planning model is adaptive, due to the use of the learned driving behavior control parameters.

The invention relates to the technical field of laser control, in particular to an intelligent control method and system of an MOPA laser. The method comprises the following steps: constructing a target training data set and a one-dimensional convolutional neural network model; training a one-dimensional convolutional neural network model based on the target training data set in combination with a target loss function and a back propagation algorithm to obtain a target model; acquiring current operation parameters of the MOPA laser; inputting the current operation parameters into a target model, and outputting target regulation and control parameters by the target model; the target regulation and control parameters comprise the working current of a pumping module in the MOPA laser and the working temperature of a doped fiber; based on the target regulation and control parameters, the working current of a pumping module in the MOPA laser and the working temperature of a doped optical fiber are adjusted, the working temperature of the doped optical fiber comprises a plurality of different target temperatures, and each target temperature corresponds to a different position on the doped optical fiber. In this way, the problems caused by the nonlinear effect generated under high-power pumping can be reduced.

The application relates to the technical field of text label marking, and provides a text specification marking method and system based on a rule corpus, which comprises the following steps: analyzing a policy and regulation document, identifying and marking condition morphemes and conclusion morphemes in the policy and regulation document, constructing a logical relationship between the two by using a large language model, and forming a rule corpus composed of structured morpheme pairs; performing semantic embedding on the corpus to generate a semantic vector library; performing multi-label coding on a verification data set based on the rule corpus, and constructing a multi-label training data set; training a deep learning classification model by taking semantic vectors as features and multi-labels as targets, so that a text specification marking model is obtained; and automatically marking target text by using the model. The application significantly improves the accuracy, interpretability and business adaptability of text marking, improves the update quality of a system label data set, and reduces the system maintenance cost.

The invention discloses a wind-solar-water-storage complementary system short-term risk scheduling method considering uncertainty, and the method comprises the steps: converging historical physical operation data and multi-subject behavior data, and constructing a training data set and a system parameter set; based on the training data set, constructing a combined robust radius and wind-solar combined scene containing behavior risk quantification; based on the system parameter set and the training data set, constructing a dynamic risk scheduling unit fusing multi-dimensional risks; solving a candidate short-term scheduling scheme by combining a joint robust radius and a dynamic risk scheduling unit and adopting a behavior risk-oriented Bayesian optimization method; and performing multi-subject consensus evaluation on the candidate short-term scheduling scheme, determining a final execution short-term scheduling scheme, and performing uplink execution. According to the method, the problem of separation of physical risks and behavior risks is solved, and the behavior acceptability of the scheme is improved.

The invention discloses a GFRP durability evaluation method based on a neural network, and belongs to the technical field of composite material performance evaluation. The method comprises the following steps: constructing a training data set containing working condition parameters, macroscopic performance data and microscopic mechanism data; a multi-head physical guidance neural network model is constructed, and the model is provided with a main output head for outputting a macroscopic performance data predicted value and an auxiliary output head which is connected to an internal physical mechanism characterization layer of the model and is used for outputting a microscopic mechanism data predicted value; performing multi-task cooperative training on the model through a composite loss function; and finally, synchronously outputting a macroscopic performance prediction result and a microcosmic mechanism diagnosis result by utilizing the trained model. According to the method, physical mechanism knowledge is fused into the neural network, and a traditional black box model is improved into an interpretable grey box model through middle layer supervision and multi-task cooperative training, so that the accuracy of macroscopic prediction is improved, and quantitative diagnosis of an internal degradation mechanism is realized.

The invention belongs to the technical field of remote sensing, and relates to a multispectral remote sensing atmospheric correction method based on fusion of a physical model and a neural network, which comprises the following steps: S1, classifying aerosol parameter data obtained based on foundation observation, remote sensing inversion or meteorological model and satellite data fusion inversion by adopting an unsupervised clustering algorithm, extracting a typical aerosol mode with physical representativeness; s2, in combination with the typical aerosol mode, simulating the apparent reflectivity of an observation channel under a plurality of atmospheric states and observation geometric conditions by using a radiation transfer model, generating a lookup table covering a wide parameter space, and constructing a training data set; and S3, constructing a nonlinear regression model, taking the training data set as a training sample, learning a mapping relation among an observation angle, an aerosol condition and surface reflectance, performing atmospheric interference correction on an actual multispectral remote sensing image under a pollution condition, and outputting a surface reflectance result.

The invention relates to a semi-supervised learning data anomaly intelligent identification and treatment system, which is applied to a hydrogen energy commercial vehicle, and comprises a data acquisition and preprocessing module, which is arranged on a vehicle-mounted terminal and is used for acquiring a hydrogen storage system signal, a hydrogen supply system signal, a fuel cell system signal and a whole vehicle system signal in real time, processing the acquired signal data; the semi-supervised anomaly recognition module is arranged on a cloud platform, is connected with the data acquisition and preprocessing module, and is used for fusing the processed signal data into a rule engine and semi-supervised learning, constructing a semi-supervised training data set, and performing deep auto-encoder model training through the data set so as to realize accurate anomaly recognition of few sample scene writing; and the exception treatment and feedback module is arranged at an edge node, is connected with the semi-supervised exception recognition module, carries out exception recognition through a trained model, carries out graded treatment according to the exception severity, and establishes a model evolution mechanism to realize continuous optimization of the system.