Patents
Literature
Patsnap Eureka AI that helps you search prior art, draft patents, and assess FTO risks, powered by patent and scientific literature data.

961 results about "Hyperparameter optimization" patented technology

In machine learning, hyperparameter optimization or tuning is the problem of choosing a set of optimal hyperparameters for a learning algorithm. A hyperparameter is a parameter whose value is used to control the learning process. By contrast, the values of other parameters (typically node weights) are learned.

Multi-layer semantic understanding large model agent construction and application method

The invention discloses a multi-layer semantic understanding large model agent construction and application method. The method comprises the following steps: S1, preprocessing input data; s2, analyzing the preprocessed data by using a large language model and dialogue history, identifying user intentions and extracting deep semantic features; s3, key information is extracted from data intending to be recognized, structured representation is carried out, and model configuration is adjusted through dynamic hyper-parameter optimization; s4, querying the knowledge graph, associating and organizing the extracted key information with the existing data, and generating a structured knowledge representation; s5, vectorizing the structured knowledge representation, and storing the vectorized structured knowledge representation in a vector database to support rapid similarity search; and S6, in combination with the generated unified semantic representation and the large model, performing deep analysis on user input, generating coherent and logical personalized answers, and returning the personalized answers to the user. The method can support various application scenes such as intelligent question and answer, content generation, sentiment analysis and multi-modal search.
Owner:SHANGHAI ENTROPY INFORMATION TECH CO LTD

Adaptive management system for IoT networks utilizing dynamic fuzzy logic framework

A system is provided for managing Internet of Things (IoT) networks. The system includes a learning module configured to employ machine learning models with hyperparameters optimized through a hyperparameter optimization process; wherein the process includes evaluating a set of hyperparameters against a performance metric to select optimal hyperparameters that enhance the adaptability and efficiency of dynamic membership functions within an adaptive fuzzy logic engine (AFLE).
Owner:LEPTUDE INC

Automatic driving large model training optimization method based on multi-scene data balance

The invention relates to an automatic driving large model training optimization method based on multi-scene data balance. Comprising the following steps: (1) constructing a real vehicle high-speed driving scene library; (2) constructing a visual language automatic driving large model, training by adopting an iterative training framework based on a real vehicle high-speed driving scene library, designing a multi-task joint loss function and a weight adaptive adjustment strategy, realizing multi-task target balance, and obtaining a trained visual language automatic driving large model; (3) dynamic simulation is carried out for the automatic driving working condition, and a high-fidelity simulation data test set is constructed based on simulation data; and (4) performing hyper-parameter optimization and lightweight processing on the trained visual language automatic driving large model according to the high-fidelity simulation data test set to obtain a scene data balanced automatic driving large model. According to the method, the large model reasoning speed is increased, and resource occupation is reduced, so that the judgment capability of the large model on dynamic working conditions and high-risk scenes is remarkably improved.
Owner:NANJING UNIV OF SCI & TECH

Magnetorheological damper modeling method based on Gaussian process regression model

The invention discloses a magneto-rheological damper modeling method based on a Gaussian process regression model. The magneto-rheological damper modeling method comprises the following steps: step 1, acquiring and preprocessing data based on a mechanical property test of a magneto-rheological damper; 2, determining Gaussian process regression prior distribution of the magnetorheological damper; 3, calculating a posteriori mean value and a posteriori covariance of the Gaussian process regression model; 4, evaluating the Gaussian process regression model of the magnetorheological damper; 5, performing hyper-parameter optimization calculation; and step 6, final evaluation and application of the Gaussian process regression model of the magnetorheological damper. On the premise of ensuring the model precision, the problems of insufficient computing resources, increased training time cost and the like possibly caused when a non-parameterized model represented by a neural network model faces a large-scale data set are solved; meanwhile, for the problem of model overfitting possibly caused by noise interference in training data, a kernel function with a noise interference suppression effect is provided so as to improve the modeling precision of the model.
Owner:ZHEJIANG SCI-TECH UNIV

Hyperparameter optimization with operational constraints

Hyperparameters for tuning a machine learning system may be optimized using Bayesian optimization with constraints. The hyperparameter optimization may be performed for a received training set and received constraints. Respective probabilistic models for the machine learning system and constraint functions may be initialized, then hyperparameter optimization may include iteratively identifying respective values for hyperparameters using analysis of the respective models performed using an acquisition function implementing entropy search on the respective models, training the machine learning system using the identified values to determine measures of accuracy and constraint metrics, and updating the respective models using the determined measures.
Owner:AMAZON TECH INC

Spinel multi-objective reverse design method based on machine learning and Bayesian optimization algorithm, electronic equipment and storage medium

The invention relates to a spinel multi-objective reverse design method based on machine learning and a Bayesian optimization algorithm, electronic equipment and a storage medium, and the design method comprises the steps: firstly extracting related data of a spinel material from a database, and constructing a balanced data set through preprocessing; feature engineering is carried out, a comprehensive feature set is constructed, and key features are reserved; then training a multi-target prediction model through hyper-parameter optimization by using a multi-task gradient elevator model; and finally, integrating to a Bayesian reverse design framework, expanding a design space through a specific encoder, and combining a Gaussian process proxy function and an expected hyper-volume improvement criterion to screen candidate materials meeting conditions and verify performance, thereby completing reverse design optimization. Compared with the prior art, the intelligent and efficient spinel novel multi-target reverse design method can be used for solving the problem of data scarcity, and development of high-performance spinel solar cell materials is accelerated.
Owner:SHANGHAI UNIVERSITY OF ELECTRIC POWER

Marine diesel engine vibration fault early warning method based on improved VMD

The invention discloses a marine diesel engine vibration fault early warning method based on an improved VMD, and the method comprises the steps: improving the VMD through combining a frost ice algorithm RIME, carrying out the hyper-parameter optimization, carrying out the VMD signal decomposition of an optimized data set based on an optimal combination, and obtaining a plurality of IMF values, so as to reconstruct vibration signal sequence data. Performing time domain and frequency domain feature extraction on the reconstructed vibration signal sequence data to obtain a time domain feature parameter and a frequency domain feature parameter; calculating and acquiring a Pearson coefficient between each feature parameter in the mixed domain feature set and the early warning value, and selecting the feature parameter of which the Pearson coefficient meets a preset Pearson coefficient threshold as an optimized feature parameter; and obtaining a residual error between a predicted root-mean-square value and an actual root-mean-square value based on the optimal BiGRU algorithm model so as to obtain a residual error threshold range under the condition of fault-free operation of the marine diesel engine, building an RIME-VMD-BIGRU model for vibration fault early warning of the marine diesel engine, and realizing automatic prediction of the vibration fault of the marine diesel engine. According to the method, sailors can be reminded of possible faults of the ship, the dependence on expert experience is reduced, and meanwhile, compared with other early warning algorithms, the method has higher diagnosis precision.
Owner:DALIAN MARITIME UNIVERSITY

Adaptive feature weighted fusion fault diagnosis method for multi-source heterogeneous data of motor

The invention relates to an adaptive feature weighted fusion fault diagnosis method for multi-source heterogeneous data of a motor, and the method comprises the steps: constructing a multi-head deep convolutional neural network, designing a plurality of parallel convolutional channels in the neural network, deeply mining the abnormal features of original signals of current and vibration, and focusing on the signal features in different dimensions, thereby achieving the fault diagnosis of the multi-source heterogeneous data of the motor. And comprehensive capture of key feature information is ensured. And adaptive weighted fusion is performed on the extracted multi-dimensional features by adopting a correlation analysis method to highlight abnormal key features, so that the feature fusion effect of the multi-source heterogeneous data and the accuracy of anomaly detection are improved. An SELU activation function is adopted and a hyper-parameter optimization strategy of a dynamic attenuation learning rate is combined so that an overfitting problem which is easy to occur in a model training process can be effectively solved and convergence stability of the model and high precision of fault identification can be enhanced. Through adaptive fusion of multi-source heterogeneous data features, key fault features are effectively extracted, and the diagnosis accuracy in an electromechanical coupling fault scene is improved.
Owner:ZHENGZHOU UNIVERSITY OF LIGHT INDUSTRY

System And Method For Dynamic Hyperparameter Optimization For Large Language Models Using (Few-Shot) Reinforcement Learning

Techniques for increasing the quality of output from large language models using reinforcement learning to select inference-time hyperparameters are disclosed. The large language model is configured with a set of values corresponding to a set of inference-time hyperparameters that are used to influence the output of the machine learning model after the model has been frozen. After obtaining a set of performance metrics that indicate the quality of the output, a reinforcement learning agent computes an adjustment for one or more of the hyperparameters, resulting in a modification of the hyperparameter values. Applying the new hyperparameter values, the large language model is then applied to a new set of input to generate a second output. The process iterates until the performance metrics associated with the output are satisfactory.
Owner:ORACLE INT CORP

Typhoon rapid enhancement prediction method based on time-space sequence and multi-modal feature fusion

The invention relates to the technical field of typhoon prediction, and discloses a typhoon rapid enhancement prediction method based on time-space sequence and multi-modal feature fusion, and the method comprises the steps: constructing a multi-modal time-space sequence data set and an auxiliary data set based on typhoon optimal path data and multi-source satellite observation data; a unified manifold approximation and projection method is adopted to carry out dimension reduction preprocessing on the high-dimensional multi-modal space-time sequence data, and one-dimensional time sequence embedding representation of the typhoon observation sequence is generated; taking the one-dimensional time sequence embedded representation and the auxiliary data as independent input channels, and inputting a trained typhoon observation network model to predict a typhoon rapid enhancement probability; wherein the typhoon observation network model is a multi-mode time-space fusion deep learning architecture, the core of the typhoon observation network model is composed of a variational attention recurrent neural network, and hyper-parameter optimization is carried out through an improved Harris eagle optimization algorithm. According to the invention, accurate and robust identification of the typhoon rapid enhancement process is realized.
Owner:NATIONAL METEOROLOGICAL CENTRE

Multi-target intelligent optimization method and system for blasting parameters of strip mine in high-altitude cold region

The invention discloses a multi-target intelligent optimization method and system for blasting parameters of a strip mine in a high-altitude cold region. The method comprises the following steps: carrying out data acquisition to obtain a parameter data set; performing data preprocessing on the parameter data set to obtain a feature sample set; constructing an initial blasting parameter model based on a machine learning algorithm, and performing hyper-parameter optimization on the model to obtain a blasting parameter model; a multi-objective optimization function is constructed: based on the multi-objective optimization function and the blasting parameter model, solving is carried out in combination with environmental condition constraints, and a pareto optimal solution set is obtained; according to the pareto optimal solution set, a representative solution is selected, a visual scheme is generated, and blasting parameter optimization of the strip mine in the high-altitude cold region is completed. According to the method, temperature, oxygen and frozen soil constraint conditions of the high-cold and high-altitude environment are introduced, blasting safety, lumpiness uniformity and the explosive utilization rate are considered at the same time through multi-target collaborative optimization, the method can adapt to the extreme environment, meanwhile, the one-sidedness of single-target optimization is avoided, and the intelligent level of blasting design and implementation is greatly improved.
Owner:CINF ENG CO LTD

Cardiovascular disease prediction model interpretable method based on SHAP

The invention discloses an explainable method of a cardiovascular disease prediction model based on SHAP, and aims to construct an accurate cardiovascular disease prediction model and provide explainable analysis by combining multi-source medical data. The method comprises the following steps: acquiring medical data related to cardiovascular diseases, performing data preprocessing and feature engineering coding, and establishing a cardiovascular disease data set; carrying out importance evaluation on the features by using a Boruta algorithm, screening out features which play an important role in cardiovascular disease prediction in combination with a random forest, and completing the division of a training set and a test set; adopting a grid search algorithm to adjust and optimize hyper-parameters, and optimizing the performance of the model; training the training set data by using an XGBoost algorithm, establishing a cardiovascular disease prediction model, and outputting feature importance; and inputting the data set into the trained model, outputting a prediction result of each patient, visualizing the prediction results by using an SHAP interpretation tool, and analyzing the interaction effect of characteristic variables and the contribution of a single characteristic. According to the technical method, accurate prediction and analysis support can be provided for clinic, doctors are helped to formulate effective prevention measures, and therefore the early diagnosis and treatment effect of cardiovascular diseases is improved.
Owner:李锦涛

Meteorological prediction precision improvement method and device, equipment, storage medium and computer program product

The invention relates to the technical field of meteorological prediction, in particular to a meteorological prediction precision improvement method and device, equipment, a storage medium and a computer program product. The method comprises the following steps: constructing a plurality of initial weather prediction models based on a Transform structure in combination with a PIDL (Precision Independent Design Language) algorithm; taking the power curve of the wind turbine generator and the performance characteristic curve of the photovoltaic module as constraint conditions, and integrating the constraint conditions into each initial meteorological prediction model to obtain a plurality of basic meteorological prediction models; based on a preset automatic hyper-parameter tuning tool, performing hyper-parameter optimization on each basic meteorological prediction model in combination with a cross validation algorithm; fusing the plurality of basic meteorological prediction models after hyper-parameter optimization by adopting an integrated learning algorithm to obtain a target meteorological prediction model; and according to the meteorological characteristic data and the target meteorological prediction model, the target meteorological prediction result is determined, and the meteorological prediction precision is improved.
Owner:济南作为科技有限公司

Risk prediction and adaptive budget allocation system and method based on construction settlement

The invention discloses a risk prediction and adaptive budget allocation system and method based on construction settlement, and the method comprises the following steps: S1, collecting historical settlement and initial budget data of a construction project, carrying out the multi-dimensional marking, and generating a structured data set; s2, processing the data by applying multi-level time sequence analysis, and identifying key levels; s3, generating an optimal quantity-price coupling relationship based on a multi-party game model, and performing preliminary tuning; s4, performing hyper-parameter optimization through a dual-domain heterogeneous population tuning mechanism; s5, analyzing the influence of external factors on the cost by using a structural transition model, and carrying out deviation analysis; s6, risk scenes under different construction situations are generated through risk simulation, and risk factors are identified and adjusted; s7, adjusting budget allocation in real time based on an adaptive complex network mechanism; and S8, a review report is generated, and multi-party collaborative review is supported. According to the method, the dynamic adjustment capability of budget allocation is improved through various optimization mechanisms, and audit transparency is improved.
Owner:STATE GRID JIANGSU ECONOMIC RES INST

Cable state evaluation method based on Bayesian optimization and XGBoost

The invention discloses a cable state evaluation method based on Bayesian optimization and XGBoost, and the method employs a cross entropy loss function on the basis of an XGBoost algorithm, and can effectively reduce the loss. And hyper-parameters in the XGBoost algorithm are adjusted and optimized by using a Bayesian parameter optimization method, so that key features in the operation process of the cable are extracted more effectively, and finally, the state of the cable is evaluated. By using the cross entropy loss function, the complexity of the cable state can be better adapted, and the robustness of the model is improved. Meanwhile, the Bayesian hyper-parameter optimization is beneficial to finding the optimal model parameter configuration, so that the model can capture the cable state characteristics more accurately.
Owner:STATE GRID HENAN INFORMATION & TELECOMM CO

Distributed optical fiber sensing perimeter intrusion event open set identification method based on self-supervised contrast learning

A distributed optical fiber sensing perimeter intrusion event open set identification method based on self-supervised comparative learning comprises the following steps: step 1, an experiment scene: establishing a DAS system, and collecting data samples of various intrusion events; 2, data preprocessing: carrying out noise reduction processing on the experimental data sample, and then converting the experimental data sample into a GASF image; step 3, the processed data is divided into two parts during training, one part is data with labels and known categories, and the other part is data without labels and containing all types of events; step 4, constructing a self-supervised contrast open set class learning (SCOC) algorithm model; step 5, training the constructed SCOC algorithm model, performing hyper-parameter optimization by using a BoHo optimization algorithm, and obtaining an SCOC algorithm training model under the optimal hyper-parameter combination; and step 6, open set event classification testing. According to the invention, the problems of identification and alarm of unknown intrusion events in an open set scene and the like can be solved.
Owner:ARTIFICIAL INTELLIGENCE INNOVATION RES INST OF ZHEJIANG UNIV OF TECH BINJIANG DISTRICT HANGZHOU

Laser welding spatter real-time identification method based on OCT keyhole depth measurement

The invention discloses a laser welding spatter real-time recognition method based on OCT keyhole depth measurement. The method comprises the steps of S1, keyhole depth data collection based on OCT measurement and data set construction and S2, construction training of a deep learning model and spatter defect recognition. According to the method, an SD-OCT measurement technology is combined with a composite deep learning model based on 1DCNN-BiLSTM splash defect recognition, a Dropout technology and adaptive moment estimation are adopted to prevent model overfitting, and a TPE algorithm is adopted to carry out hyper-parameter optimization on the model, so that the trained 1DCNN-BiLSTM splash defect recognition model can accurately predict the splash state in the dynamic laser welding process, and the accuracy of the dynamic laser welding process is improved. Manual intervention is not needed, and the laser welding spatter defect can be rapidly and accurately recognized.
Owner:TAIER WISDOM (SHANGHAI) LASER TECH CO LTD +1

Target trajectory prediction method and system based on bidirectional long and short term memory network

The invention relates to a target trajectory prediction method and system based on a bidirectional long-short-term memory network, and the method mainly comprises the following steps: firstly, carrying out the spatial-temporal feature extraction of target trajectory data through a bidirectional long-short-term memory network layer, obtaining the spatial-temporal feature vector of target data, capturing a long-term dependency relationship and a short-term dependency relationship, and carrying out the spatial-temporal feature extraction of the target trajectory data; wherein the long-term dependency describes the time sequence and continuity of motion, and the short-term dependency describes the maneuvering characteristics of the motion; secondly, a prediction model is constructed, and hyper-parameters of the prediction model are optimized, so that the prediction performance and generalization ability of the model are improved; and finally, predicting a future trajectory of the target by using the trained optimization model, outputting a trajectory coordinate sequence of a plurality of time points in the future, and realizing accurate prediction of the target in a period of time in the future. According to the method, the bidirectional long-short-term memory network and hyper-parameter optimization are combined, so that high-precision trajectory prediction of the target is realized, and the adaptability of the prediction model to complex motion characteristics is enhanced.
Owner:XIAN TECH UNIV

Data-driven modeling method for machine tool thermal error caused by environment temperature change

The invention belongs to the field of precision numerical control machine tool error compensation, and discloses a data-driven modeling method for machine tool thermal errors caused by environment temperature change, which comprises the following steps of: analyzing a forming mechanism of the machine tool thermal errors caused by the environment temperature change; establishing a data driving sub-model of environment temperature variation-machine tool surface temperature variation and a data driving sub-model of machine tool surface temperature variation-machine tool internal temperature variation-machine tool thermal deformation; performing hyper-parameter optimization on the data driving sub-models, and connecting the two data driving sub-models in series to form an environment temperature-machine tool thermal error data driving model. According to the method, a machine tool thermal error forming mechanism is analyzed, a thermal error neural network structure with a memory function is designed, model parameters are identified by utilizing multi-physical-quantity data such as environment temperature, machine tool surface temperature and machine tool thermal errors, and the interpretability and generalization ability of a data driving model are improved; the thermal error is predicted only through one environment temperature data, and the method is simple and efficient.
Owner:IND TECH RES INST OF YIBIN SICHUAN UNIV

Load characteristic prediction method based on Informer hyper-parameter optimization model

The invention relates to the technical field of power system load prediction, and discloses a load characteristic prediction method based on an Informer hyper-parameter optimization model, which comprises the following steps of: firstly, acquiring long-time weather and power grid data of a power supply company region and preprocessing to obtain a characteristic data set corresponding to load data; and carrying out normalization processing on the feature data set corresponding to the load data. Then, inputting the processed data into an Informer model for training, fully exploring the time correlation among the data, and utilizing a multi-head sparse self-attention mechanism to mine the correlation between the states before and after the feature information; and then determining hyper-parameters needing to be optimized and a value range, and performing hyper-parameter optimization on the Informer model by adopting a TPE algorithm to obtain an optimal combination of the hyper-parameters. And finally, obtaining a prediction result through the optimized model, then establishing a model comparison scheme, and carrying out prediction result comparative analysis on the Informer parameter optimization model, the LSTM and the GRU to embody the feasibility of the method.
Owner:DAZHOU POWER BUREAU SICHUAN ELECTRIC POWER

Settlement time sequence prediction method and system for deep foundation pit excavation adjacent building

The invention discloses a settlement amount time sequence prediction method and system for deep foundation pit excavation adjacent buildings. The method comprises the steps that original monitoring data of on-site building settlement are acquired; performing data preprocessing on the obtained original monitoring data; constructing a recurrent neural network; model input and output parameters are determined through principal component analysis; optimizing the recurrent neural network based on an optimizer; performing hyper-parameter optimization based on an optimization result; performing settlement time sequence prediction by using the optimized recurrent neural network; the system comprises a data acquisition module, a preprocessing module, a model construction module, an analysis module, an optimization module, a parameter optimization module and a prediction module. By constructing the settlement prediction model, dynamic modeling and accurate prediction of the settlement trend of the building in the excavation process of each stage of the foundation pit are realized; historical settlement monitoring data and multi-layer soil body excavation information are combined, and multi-source input parameters are introduced, so that the adaptability of the model to complex working conditions is enhanced.
Owner:SHANDONG JIANZHU UNIV

Construction parameter optimization method and device based on dessert prediction, medium and equipment

The invention discloses a construction parameter optimization method and device based on dessert prediction, a medium and equipment, and the method comprises the steps: obtaining a data set of a shale gas well production process, and determining a dessert indication factor; training a network model by taking the data set as input and the dessert indication factor as output to obtain a dessert indication factor prediction model; optimizing the dessert indicator factor prediction model by using a Bayesian hyper-parameter optimization algorithm, and inputting a data set obtained in real time into the optimized dessert indicator factor prediction model to obtain a prediction result; inverting a prediction result into three-dimensional space distribution characteristics of desserts in the shale reservoir; and determining contribution values and partial dependency relationships of the plurality of geological parameters and the plurality of engineering parameters to a prediction result based on the three-dimensional space distribution characteristics, deploying construction parameters according to the contribution values and the partial dependency relationships, and synchronously updating a sweet spot indication factor prediction result to realize spatial dynamic optimization and adjustment of the construction scheme. Through the scheme, the construction parameters can be scientifically optimized.
Owner:NORTHEAST GASOLINEEUM UNIV

High-precision single cell classification method and device based on artificial intelligence algorithm

The invention discloses a high-precision single cell classification method and device based on an artificial intelligence algorithm, and the method comprises the steps: firstly, converting original single cell data into a csv format, and constructing an expression matrix to describe the expression levels of genes under different conditions; thirdly, preliminarily dividing single cell categories through dimensionality reduction and clustering analysis; further, in combination with PPI and GO information, a graph neural network method is used for carrying out quantitative characterization and classification on the genes; then, constructing a single cell data embedding model, extracting features through an encoder and a decoder, and training the model through cross validation and hyper-parameter optimization; and finally, a hidden layer of an encoder in the model is used as an embedded expression matrix of the single cells, the layer comprises original single cell expression information, PPI, GO and other multi-dimensional data, and various downstream tasks including single cell accurate classification can be realized. The method is helpful for deeply understanding a gene expression mode, and provides data support for disease treatment and drug design.
Owner:ZHEJIANG LAB

Space-time carbon emission prediction method, system, equipment and medium

The invention discloses a space-time carbon emission prediction method, system and device and a medium. The method comprises the steps of obtaining load historical sequence data; carrying out load flow calculation by adopting a first load flow analysis algorithm to obtain a load flow calculation result; based on the load flow calculation result, carbon flow analysis is carried out through power transmission between the nodes, and the carbon emission coefficient of each node is obtained; constructing a time sequence diagram neural network model, and performing preliminary training on the time sequence diagram neural network model; performing hyper-parameter optimization on the preliminarily trained time sequence diagram neural network model by adopting a first hyper-parameter optimization technology to obtain an optimized time sequence diagram neural network model; and inputting load sequence data and unit power output data at the current moment into the optimized sequence diagram neural network model, and carrying out online prediction on a carbon emission coefficient. According to the method, the real-time prediction of the carbon emission coefficient is realized based on the characteristic learning of the historical operation and monitoring data of the power system.
Owner:YUNNAN POWER GRID CO LTD

Short-term load prediction method

The invention provides a short-term load prediction method, which comprises the following steps: collecting multivariable data of a power system, and preprocessing the collected multivariable data; wherein the multivariable data at least comprises power load data, weather data, holiday and festival information and date data; inputting the preprocessed multivariable data into a pre-trained prediction model to obtain a corresponding prediction result; wherein the prediction model is a model combining a time convolution network and a long short-term memory network, hyper-parameter optimization is carried out through a preset search algorithm, training is carried out through historical multivariable data, and a plurality of intrinsic mode component prediction results are combined to be output as the model; and determining the short-term load corresponding to the power system according to the prediction result. According to the method, the nonlinear features in the power load data are effectively extracted, more accurate load prediction is realized through advantage complementation of multiple models, and a solid guarantee is provided for stable operation of a power system.
Owner:SHENZHEN POWER SUPPLY BUREAU

Coal gangue concrete stress response prediction method based on physical information neural network

The invention discloses a coal gangue concrete stress response prediction method based on a physical information neural network. The method comprises the following steps: constructing a stress-strain curve database of coal gangue concrete; acquiring damage variables of the coal gangue concrete in a compaction stage, a yield stage and a fracture stage under uniaxial compression; deducing a mesoscopic statistical damage constitutive model of the coal gangue concrete according to the damage variables: constructing a physical information neural network, and embedding the mesoscopic statistical damage constitutive model and boundary conditions into a loss function of the neural network; performing normalization processing on the coal gangue concrete stress-strain curve database to obtain a data set; training a physical information neural network model according to the training set, and performing hyper-parameter optimization according to the verification set to obtain a trained model; and predicting the stress of the coal gangue concrete by using the trained model according to the test set. According to the method, the damage constitutive equation and the deep learning algorithm are fused, so that high-precision and low-data-demand mechanical property prediction is realized.
Owner:CHONGQING UNIV

Tilting type fuel gas aluminum melting furnace comprehensive energy consumption prediction method based on CPO-ITCN-GRU

The invention provides a comprehensive energy consumption prediction method for a tilting type fuel gas aluminum melting furnace based on CPO-ITCN-GRU. The method comprises the steps that original energy consumption data generated in the running process of the tilting type fuel gas aluminum melting furnace and auxiliary characteristic data related to energy consumption are collected; comprehensive energy consumption conversion is carried out on the original energy consumption data, and denoising smoothing is carried out on the comprehensive energy consumption data obtained through conversion and auxiliary feature data related to energy consumption; dividing the denoised and smoothed data into a training set and a test set according to a preset proportion; constructing an ITCN-GRU comprehensive energy consumption prediction model, training the ITCN-GRU comprehensive energy consumption prediction model by using the training set, and performing hyper-parameter optimization by using a crown porcupine optimization algorithm in the training process; and performing performance evaluation on the trained ITCN-GRU comprehensive energy consumption prediction model by using the test set. According to the method, the capturing capability of the model on the local space-time characteristics and accurate hyper-parameter optimization are improved, so that the comprehensive energy consumption prediction accuracy of the tilting type fuel gas aluminum melting furnace is improved.
Owner:ZHENGZHOU UNIVERSITY OF LIGHT INDUSTRY

Metalearning-based hyper-parameter optimization method and system for icing galloping prediction model

The invention discloses a hyper-parameter optimization method and system of an icing galloping prediction model based on meta-learning. The method comprises the following steps: collecting original data from meteorological monitoring equipment, a line state sensor and historical icing galloping records; uniformly aligning the collected original data according to time granularity to obtain multi-modal time sequence data; performing internal and external double-loop training on hyper-parameters of the constructed icing galloping time sequence prediction model by adopting the obtained multi-modal time sequence data, performing iterative training on internal loops of the internal and external double-loop training under the given hyper-parameters, and after model parameters converge, obtaining the icing galloping time sequence prediction model. And performing meta-learning-level hyper-parameter optimization on the basis of converged model parameters in the outer loop, and performing adaptive search on hyper-parameters on the premise of fixing an inner loop training normal form to obtain optimal hyper-parameters. According to the method, the key hyper-parameters of the model can be quickly and adaptively optimized under different meteorological scenes and line working conditions, the prediction precision and the convergence speed are improved, and the icing galloping risk of the power transmission line is warned in real time.
Owner:STATE GRID JIANGSU ELECTRIC POWER CO LTD +2

Auto tuned swarm learning models

Swarm learning systems and methods can perform hyperparameter tuning. Network nodes train a common machine learning (ML) algorithm using data local to each of the network nodes. Hyperparameter optimization is also performed at each of the network nodes. Model parameters and optimized hyperparameters are shared by the network nodes, and are subsequently merged. An updated version of the common ML algorithm based on the merged model parameters and the merged optimized hyperparameters is generated. That updated version of the common ML algorithm is re-trained at each of the network nodes.
Owner:HEWLETT PACKARD ENTERPRISE DEV LP

Ultra-short-term wind power prediction method based on Bayesian optimization XGboost-LSTM

The invention belongs to the technical field of wind power prediction, and discloses an ultra-short-term wind power prediction method based on Bayesian optimization XGboost-LSTM, and the method comprises the following specific steps: 1, inputting wind power data collected by a wind power plant and numerical weather prediction data of a corresponding time sequence; 2, data preprocessing, wherein missing value interpolation and data deduplication are carried out on input data; according to the method, an XGBoost feature optimization method is used for selecting key features influencing the wind power, the influence of irrelevant feature noise on the model prediction precision and accuracy is eliminated, then a Bayesian optimization algorithm is used for carrying out hyper-parameter tuning on an LSTM model, and finally an XGBoost-LSTM-BO model is constructed. XGboost feature optimization and Bayesian hyper-parameter optimization can obviously improve the prediction effect of the LSTM on future data and improve the model prediction precision, and compared with a traditional prediction model, the wind power data generalization ability of the model can be improved while the high prediction precision is kept, and higher prediction performance is achieved.
Owner:INNER MONGOLIA UNIV OF TECH