125results about How to "Improve model performance" patented technology

The invention discloses a high-speed rail system sliding layer peak displacement envelope curve and rail direction irregularity prediction method considering seismic oscillation and pier height uncertainty and application. The method comprises the following steps: constructing a multi-mode hybrid neural network model TransTCN embedded with field priori knowledge; feature extraction is carried out on multi-modal input features of a time sequence based on seismic oscillation fusion structure features and a space sequence for outputting peak displacement based on an elastic response spectrum method; the pier height features and the seismic oscillation features are subjected to dimension fusion to serve as time sequence features, and a time dependency relationship is processed; processing a spatial dependency relationship by taking an elastic response spectrum analysis result as a spatial feature; and jointly inputting the extracted features into a constructed multi-mode hybrid neural network model TransTCN, and realizing prediction of the peak displacement envelope curve and rail direction irregularity of the sliding layer of the high-speed rail system under the consideration of seismic oscillation and pier height uncertainty. The prediction precision is improved, the training sample demand is reduced, the regional high-speed rail system damage is rapidly evaluated, and the practical value is good.

The invention discloses an on-load tap-changer fault diagnosis method based on lightweight YOLO11, and is used for solving the problem of low mechanical fault diagnosis efficiency of a traditional on-load tap-changer. Comprising the following steps of: converting vibration signals in a motor starting stage and a change-over switch action stage into high-resolution time-frequency images through continuous wavelet transform (CWT) so as to comprehensively reflect time and frequency characteristics of the signals; starNet is adopted to replace a YOLO model backbone network, and the feature extraction capability is enhanced while the network compactness is kept through star operation; and designing an efficient joint convolution detection head EJCD, and realizing model lightweight and multi-scale target detection through grouping normalization and convolution parameter sharing. According to the invention, through combination of the time-frequency image representation capability of continuous wavelet transform and an efficient target detection framework of deep learning, the limitation of a traditional method in non-stationary signal processing is overcome, and the diagnosis efficiency and the model performance are significantly improved.

The invention relates to a marine ranch power supply system line fault diagnosis method based on MS-2DResNet and ICBAM, and belongs to the field of power supply system line fault diagnosis. The fault diagnosis method is mainly divided into two stages, namely, multi-scale feature extraction and attention mechanism feature fusion. Firstly, a preprocessed fault data sample is converted into two-dimensional data to be input into a fault diagnosis model; in the first stage, local and global feature extraction is performed on the converted fault data by using multi-scale convolution, so that the robustness of the model for fault data features of different sizes and scales is improved. And then, entering an attention feature fusion stage, and carrying out weighted fusion on stacked data from different scale channels by utilizing the improved CBAM, so that the model pays more attention to useful fault features and neglects useless features, and further, the fault diagnosis accuracy of the model is improved. The method has the characteristics of high diagnosis accuracy and good anti-noise performance.

The invention belongs to the technical field of Internet of Vehicles and communication security, and relates to a vehicle intrusion detection system and method based on a Hybrid EfficentNet model, and the system is obtained by training the Hybrid EfficentNet model. Wherein the Hybrid OfficientNet model comprises an initial convolutional layer, a backbone network, a self-adaptive global pooling layer, a Flaten layer, a discarding layer, a hidden layer and a classification layer; the backbone network is composed of one Fusion MBConv module and 15 MBConv modules, and each MBConv module comprises an expansion layer, a depth separable convolution module, a channel and space attention mechanism and a projection layer; residual connection is used in the MBConv module; the system has excellent capability of detecting abnormal traffic, realizes high accuracy, high-efficiency processing capability and high response speed, meets the requirement of convenient operation in a vehicle environment, and effectively solves the technical problems that an existing intrusion detection model is complex and inconvenient and cannot be directly deployed in a resource-limited vehicle-mounted network environment.

The invention provides a subway pedestrian anomaly detection method and system based on an improved yov8 network and a clip model, and relates to the technical field of subway pedestrian detection, and the method comprises the steps: obtaining a video frame in a subway scene, and carrying out the preprocessing of the video frame; inputting the preprocessed frame data into an improved yolov8 neural network model to obtain an abnormal behavior target detection frame, and performing target candidate tracking on the target detection frame by using a target tracking algorithm to obtain a candidate image of a target area of a specific behavior; the method comprises the following steps: inputting a candidate image containing a target area into a clip model, inputting a set detection text into the clip model, mapping the detection text and the candidate image of the target area, performing text-image similarity matching, and optimizing a similarity matching threshold by adopting a Q-Learning reinforcement learning immune algorithm to obtain a similarity matching result; and the abnormal behavior detection accuracy is improved.

The invention discloses a strawberry internal quality and maturity nondestructive testing method based on an MPCD-Mama model. The strawberry internal quality and maturity nondestructive testing method comprises the following steps: acquiring a strawberry sample, acquiring a hyperspectral image by using a hyperspectral reflection imaging system, measuring SSC, PH and hardness, and marking maturity; correcting the hyperspectral image, performing spectrum extraction to form a data set, and dividing the data set into a training set, a test set and a verification set; the method comprises the following steps: constructing an MPCD-Mama model, training the MPCD-Mama model by using a training set, testing the performance of the MPCD-Mama model by using a verification set on the basis of a cross validation method, predicting related indexes on the basis of the pre-trained MPCD-Mama model, generating heat maps of corresponding components at different positions of strawberries, and displaying the distribution conditions of the different components. According to the method, nondestructive testing of multiple quality parameters of strawberries is achieved, compared with a traditional method, the prediction precision is remarkably improved, and technical support is provided for nondestructive testing of the quality of the strawberries.

The invention relates to an ECG anomaly detection method based on AE-GAN, and belongs to the field of artificial intelligence. According to the method, AE-GAN is constructed and trained for ECG anomaly detection, and a training network model is divided into an encoder training stage, the AE-GAN is only trained on a normal ECG sample, and an encoder, a generator and a discriminator are optimized by using confrontation loss and reconstruction loss; the encoder includes a channel attention module and a memory module. In a classifier training and detection stage, the AE-GAN extracts low-dimensional features from a real sample by using a trained encoder, adds weak random Gaussian noise to generate a pseudo normal sample, and adds strong random Gaussian noise to generate a pseudo abnormal sample; the classifier is trained by using a binary cross entropy loss function so as to distinguish normal samples from pseudo abnormal samples; during testing, the trained classifier uses features extracted by the encoder to detect whether data is abnormal or not.

The invention discloses an urban inland inundation prediction monitoring method, system and device and a medium thereof, and the method comprises the steps: obtaining historical ponding distribution information and a flood prevention starting plan of a target urban monitoring region based on collected rainfall prediction data and rainfall early warning levels; acquiring current flood prevention drainage facility information and drainage facility state results according to the historical ponding distribution information and the flood prevention starting plan; in the predicted urban inland inundation deduction model, performing urban inland inundation simulation calculation based on rainfall prediction data, historical ponding distribution information, a flood prevention starting plan and a corresponding drainage facility state result to obtain an urban inland inundation simulation result; generating a plurality of waterlogging and ponding monitoring units based on the urban waterlogging simulation result, and collecting corresponding waterlogging and ponding monitoring data; optimizing the urban inland inundation deduction model when the urban inland inundation simulation result and the inland inundation ponding monitoring data meet deviation judgment adjustment conditions; according to the invention, the processing efficiency of urban flood prevention and typhoon prevention emergency scheduling is improved.

The invention discloses an automatic data analysis system and method based on artificial intelligence, and relates to the technical field of data management, and the method comprises the following steps: S1, data collection and preparation; s2, data annotation and classification; s3, feature selection and extraction; s4, data standardization and normalization are carried out; s5, model selection; s6, training and optimizing the model; s7, performing model evaluation; s8, model interpretation and visualization are carried out; s9, generating an automatic report; and the system is constructed based on the steps. According to the automatic data analysis system and method based on artificial intelligence, by setting a classification and labeling model, a training optimization mechanism and a dynamic model evaluation method, the efficiency and precision of automatic data analysis are remarkably improved, and the quality of data preprocessing is improved; meanwhile, through self-adaptive hyper-parameter adjustment and optimization, the efficiency of the training process and the model performance are optimized; the dynamic evaluation mechanism ensures that the model is more accurate in performance under different data characteristics.

The invention provides a knowledge graph completion method, and aims to improve the accuracy and effectiveness of knowledge graph completion. The method comprises the following steps: firstly, obtaining and expanding triple data of a knowledge graph, and supplementing a data set by generating an inverse triple; in combination with various negative sample generation technologies, the recognition capability of the model on error links is improved, and text embedding learning is performed on entities and relationships in a triple by using a double-text encoder. A deformable adaptive attention module is provided, and potential feature capture is carried out on text features. And calculating a cosine similarity score of text embedding and adding a path reasoning gain score to obtain a triple similarity score. And text embedding is optimized by comparing and estimating a loss function by using information noise with margin, so that the complementation effect is effectively improved. And finally, completing the model by using the trained knowledge graph, and completing triple maximum similarity matching. According to the method, complex semantic association can be captured, and the accuracy and robustness of knowledge graph completion are enhanced.

The invention discloses a power transmission line unmanned aerial vehicle inspection picture defect identification method and system. The method comprises the following steps: carrying out picture resolution amplification processing on a power transmission line unmanned aerial vehicle inspection picture; inputting the processed picture into a large target component identification model, identifying large target component defects and large target components where small target component defects are located, obtaining confidence coefficients corresponding to the identified large target component defects, and retaining each identified large target component defect greater than a confidence coefficient threshold value; cutting an image of the large target part where the identified small target part defects are located, inputting the image into a small target part identification model, identifying the small target part defects, obtaining confidence coefficients corresponding to the small target part defects, and retaining each identified small target part defect greater than a confidence coefficient threshold value; and summarizing the recognized defects of the large target part and the recognized defects of the small target part, and outputting a defect recognition result. According to the method and the device, the problem of defect identification of images with high resolution and concentrated defect distribution can be solved.

The invention discloses a livestock health state monitoring method, apparatus and device, and a storage medium. The livestock health state monitoring method comprises the steps of obtaining sensor monitoring data in a livestock factory; constructing a sample data set according to the sensor monitoring data; and performing model pre-training and model tuning training on a pre-constructed monitoring network model by adopting the sensor sample data until a preset model training ending condition is met, and taking the trained monitoring network model as a livestock health monitoring model. By means of the technical scheme, the model performance of the livestock health monitoring model applied to different livestock factory areas is improved, and meanwhile the training efficiency of model training is improved.

The invention provides a federated learning-based chip manufacturing process parameter optimization method and system, and relates to the technical field of chips, and the method comprises the steps: collecting multi-factory process parameters and yield data, screening key parameters through wavelet transform and dual evaluation, and constructing a deep Q network model; ethnic group division is carried out based on process trajectory similarity, differential privacy and homomorphic encryption are adopted to protect data security, and ethnic group optimization is realized through an attention mechanism and knowledge distillation; and setting a third-level time window monitoring yield, and dynamically adjusting an optimization strategy. According to the method, the chip manufacturing yield can be effectively improved, data privacy is protected, and intelligent optimization of process parameters is realized.

The invention discloses a three-dimensional human body posture estimation method and system based on machine learning. The method comprises the following steps: constructing a posture estimation data set; training and optimizing network parameters of the feature extraction model, the feature dimension reduction model and the regression model; and extracting a feature vector of the target image by using the feature extraction model, performing feature dimension reduction by using the feature dimension reduction model, and finally inputting the feature vector after dimension reduction into the regression model to predict the three-dimensional coordinates of each key point of the human body. According to the method, the generative adversarial network based on two-stage enhancement is used for data expansion, so that the generalization ability of the model can be improved; network parameters of the feature extraction model are optimized by adopting a neural network parameter optimization algorithm based on a fusion phenomenon, so that the problem of gradient disappearance or explosion is avoided; a sparse self-expression layer is introduced between an encoder and a decoder, so that the correlation between features is reduced; and the prediction strategy of the regression model is adjusted by adopting a reverse reinforcement learning mechanism, so that the prediction accuracy is remarkably improved.

The invention relates to the technical field of neural network training, and discloses a comprehensive training method combining pre-training and staged training of a multi-caudal neural network, and the method comprises the steps: constructing a sample set; pre-training a public hidden layer of a full-connection network model through the sample set to obtain a target weight parameter; migrating the target weight parameter to a multi-caudal neural network model to obtain an initial decision model; and performing staged training on the initial decision model through the sample set to obtain a design scheme decision model. The training efficiency and the model performance of the model can be effectively improved.

The invention belongs to the field of traffic control systems, and relates to a vehicle driving track prediction system and method based on an improved graph neural network model, and the system comprises an FPGCN model and a Social LSTM model. The FPGCN model comprises a graph convolution layer, a Dropout layer, a differential pooling layer, a feature fusion module and a full connection layer; obtaining node feature vectors from graph structure data through a graph convolution layer, a Dropout layer and a differentiable pooling layer, then dividing and conquering node features, generating global features through a dense attention mechanism and principal component analysis, and generating local features through virtual filling and matrix calculation; after global features and local features are obtained, a feature fusion module is used for fusion, then the features are input into a full connection layer, and the influence score of a historical trajectory diagram on a current state diagram is calculated; integrating the preprocessed driving data and the data of which the score is calculated by the FPGCN model, and inputting the integrated data into a Social LSTM model; according to the system, while the model expression ability is maintained, the parameter quantity is remarkably reduced, the calculation logic is optimized, and the prediction accuracy is improved.

The invention discloses an industry dynamic tag identification method combining big data and ROBERTa, and belongs to the technical field of AIOT, and the method comprises the steps: collecting text data related to the industry from a plurality of big data sources through an automatic means, carrying out the cleaning and preprocessing of the collected data, preparing the data for subsequent processing, and carrying out the recognition of the industry dynamic tag. Through the big data processing technology and the RoBERTa pre-training language model, efficient and accurate label identification of industry data is realized, and powerful support is provided for industry analysis, market prediction and the like. According to the method, the universality of big data and the deep understanding ability of RoBERTa in text processing are combined, and the accuracy and efficiency of tag identification are effectively improved.

The invention discloses a hierarchical quantization and dynamic priority management-based model cache optimization method and device. The method comprises the following steps of: acquiring an importance score of each Token of a word sequence corresponding to a reasoning long text by utilizing a mask language model; constructing a Token importance scoring model, and dividing the cloud platform GPU into a head cache region, a middle cache region and a tail cache region; the service reasoning model stores the key value pairs of the first # imgabs0 # Token in a Token reasoning sequence corresponding to the reasoning long text in a head cache region in an FP16 format, stores the key value pairs of the first # imgabs1 # Token before the current processing Token in a tail cache region in the FP16 format according to an FIFO principle, and stores the key value pairs shifted out of the tail cache region in an intermediate cache region in an FP8 format; and after the intermediate cache region is full, updating the intermediate cache region by using the Token importance scoring model and the importance score. According to the method, the reasoning efficiency of the large language model in long text generation can be improved.

The invention discloses a digital twinborn vegetation environment modeling method based on a remote sensing image and a CGA rule, and belongs to the technical field of remote sensing monitoring. Comprising the following steps: S1, an image data preprocessing stage; s2, a target detection model training and tuning stage; s3, a non-dense vegetation monomer target detection and recognition stage: performing monomer tree target detection and recognition on the orthographic panchromatic image of the target area by using the trained deep neural network model to obtain a recognition and detection frame, and performing vector data conversion to obtain geographic vector data; s4, a modeling environment preprocessing stage: performing environment preprocessing on the modeling environment to obtain a basic modeling scene; and S5, a model creation and export stage: based on the geographic vector data and the basic modeling scene, creating a single vegetation model to obtain a vegetation environment model scene. The vegetation environment is restored and modeled by bypassing surveying and mapping means with relatively high hardware, software and labor cost, such as laser point cloud and oblique photography, and the method has the advantages of high feasibility and relatively low comprehensive cost.

The invention relates to a cache management method and device, computer equipment, a storage medium and a program product. The method comprises the steps that under the condition that key value caches are distributed for model layers of a target model at least once, the target number of the key value caches needed by all the model layers is obtained; wherein the target number corresponding to the model layer is the number of key value caches required by the model layer when the accumulated attention score of the attention head of the model layer reaches a target value; one attention score is generated every time the attention head based on the model layer executes the calculation task, and the accumulated attention score is an accumulated result of the attention scores after the attention head based on the model layer executes the calculation tasks for multiple times; and adjusting the allocated key value cache of each model layer according to the target number corresponding to each model layer. By adopting the method, the difference of key value cache demands of different model layers can be considered, and the key value cache can be allocated to each model layer more reasonably and accurately, so that the resource utilization rate and the model performance are improved.

The invention provides a video point location treatment method, and belongs to the technical field of video monitoring management, and the method comprises the steps: constructing and deploying a video point location treatment system management platform which is used for obtaining hardware information, point location information and video stream information of video equipment, and carrying out the standard description of the video equipment, distributing a unique identifier to each video device, wherein the point location information comprises latitude and longitude information; point location management and proofreading tasks are distributed to the user, and user permission is set; creating a point location attribute tag, and setting and managing according to a point location management verification rule; the point location governance verification rule comprises geographic information system double verification, spatial relation verification and time calibration. According to the method, precise treatment and efficient proofreading of the video point location data are realized, standardized sharing of the monitoring point location data is ensured, the quality of the video point location data is effectively improved, and the integrity, consistency and normalization of the data are ensured.

The invention discloses a near infrared spectrum abnormal sample identification method based on a high-dimensional subspace, and belongs to the technical field of near infrared spectrum analys.The technical scheme includes the steps that original spectrum data containing normal samples and abnormal samples are collected, and spectrums of a group of samples form a spectrum matrix A; preprocessing the original spectral data to obtain a preprocessed spectral matrix D; selecting a sparse subspace according to the processed spectrum matrix D; and calculating an abnormal sample according to the obtained subspace. The method has the advantages that the anomaly detection accuracy is effectively improved through the anomaly detection method based on the subspace, the related subspace is selected according to the local distribution non-uniformity, the one-dimensional variance, the local sparsity and other standards in the high-dimensional near infrared spectrum data, and the influence of the dimensionality on anomaly recognition is reduced. The abnormal value is determined in combination with the adaptive neighborhood and the threshold value, and the influence of neighborhood parameter selection on abnormal value identification is reduced.

A short-term wind power prediction method based on a Spearman-ISSA-BiLSTM combination model comprises the following steps: screening important meteorological characteristics influencing wind power through a Spearman correlation analysis method, and taking the meteorological characteristics as input of a BiLSTM model; carrying out hyper-parameter optimization on the BiLSTM model by adopting a sparrow search algorithm improved based on Logistic chaotic mapping, adaptive weight and a local search strategy, finding the most suitable model parameters, and establishing an ISSA-BiLSTM combined model; and inputting the screened meteorological characteristics and historical power data for training to obtain a final prediction result. The method is based on a Spearman correlation analysis method, an improved sparrow search algorithm (ISSA) and a bidirectional long short-term memory network (BiLSTM) combined prediction model, compared with a traditional prediction model, the prediction precision is obviously improved, and the problem that the wind power prediction precision is not high can be well solved.

The invention discloses a cross-subject cognitive load electroencephalogram signal data enhancement method based on a generative adversarial network. Source domain data and target domain calibration data are input into the SIE-GAN data enhancement method. According to the method, an encoder module is added and used for extracting features of original cognitive load electroencephalogram signals. Meanwhile, according to the data enhancement method, the setting that random noise serves as input in a traditional GAN method is abandoned, source domain data and target domain calibration data serve as input, and due to the fact that the source domain data and the target domain calibration data can provide more contextual information, the model can better adapt to the difference between different subjects, and the accuracy of data enhancement is improved. Therefore, the generalization ability of the model across subjects is improved. Verification is carried out on a cognitive load electroencephalogram signal self-acquisition data set through a plurality of deep learning models, and the result shows that the classification accuracy of cross-subject cognitive load recognition is effectively improved.

The invention discloses a large model fine tuning method based on Multi-A and Multi-B expert hybrid hybrid experts, which is characterized in that a LoRA network is embedded into a hybrid expert (MoE) architecture, and the Multi-A and Multi-B MoE are utilized to capture differences among different tasks, so that the fine tuning efficiency and comprehensive performance of a large language model in a multi-task scene are remarkably improved. The method is suitable for various large language models, the application of the large language models in different fields is promoted, and a new thought is provided for further exploring an efficient parameter fine tuning method.

The invention provides a remote sensing image desert shrub extraction method, which is characterized in that an FETNet model is obtained through improvement based on an OfficientNet-B0 baseline model, an original framework predefined in the baseline model is reserved at an encoder part, an EdgeViT module is introduced at a jump connection part of an encoder and a decoder, the global perception capability of the model is enhanced, and the algorithm is simple and easy to implement. An FEM module is introduced into a decoder part to ensure the reservation of space detail information, so that a finally generated desert shrub segmentation result is more accurate and meticulous, a preprocessed remote sensing image is input into an FETNet model, a stochastic gradient descent optimizer, a cosine annealing strategy and a cross entropy loss function are used for model training, the optimal weight of the FETNet model is obtained, and the final desert shrub segmentation result is obtained. The FETNet model is compared with a conventional model, and under the same test condition, all evaluation indexes in the remote sensing image desert shrub segmentation task obtain optimal results.

The invention discloses a super-relation mapping knowledge domain representation method and system based on dynamic fusion of structure and text information, and belongs to the technical field of mapping knowledge domain representation. The problem that rich text information in a super-relation knowledge graph is not considered in an existing super-relation knowledge graph representation method is solved. According to the method, a graph convolutional neural network, soft cue word fine tuning, a pre-training language model, Transform, comparative learning and other technologies are used to construct a super-relation knowledge graph representation method based on dynamic fusion of structure and text information. The method can enhance the representation of entities and relationships in the atlas. In addition, the efficiency problem of the negative samples in the training process is fully considered, the negative sampling method based on comparative learning is provided, and the training efficiency and the model prediction effect are improved. The method is suitable for a knowledge graph representation technology in the field of knowledge graph completion.

The invention provides a target identification method and system based on multi-source information fusion, and relates to the technical field of low-altitude target detection. The method comprises the following steps: acquiring target radar track data, interception equipment track data and position area information data; based on target radar track data, extracting a first feature in an RCS form dimension, and extracting a second feature in a motion dimension; based on the target radar track data and the track data of the monitoring equipment, determining a frequency spectrum monitoring correlation factor and regional position information features; performing feature fusion on the first feature, the second feature, the spectrum interception correlation factor and the regional position information feature to obtain a target feature, and identifying a target type; and identifying a target threat level based on the target type and the target radar track data. The method and the device are used in a target identification process based on multi-source information fusion, and the technical problem that the target threat degree cannot be accurately identified in a complex environment in the prior art is solved.

The invention relates to the technical field of reservoir parameter prediction, and provides a dolomitization parameter data prediction model establishment method, a dolomitization parameter data prediction method and a dolomitization parameter data prediction device. The dolomitization parameter data prediction model establishment method comprises the steps of obtaining a dolomitization parameter data set corresponding to a dolomitization sample; screening the dolomitization parameter data set to obtain a screening result; constructing a training sample according to the screening result; utilizing a hyper-parameter optimization algorithm to optimize hyper-parameters in the gradient boosting decision tree model; and training the optimized gradient boosting decision tree model by using the training sample to obtain a dolomitization parameter data prediction model. According to the embodiment of the invention, the evaluation precision of dolomitization parameters can be improved.

The present application belongs to the field of computer technology, and discloses a photovoltaic load prediction method and device based on artificial intelligence, the method comprising: obtaining historical photovoltaic data of a photovoltaic power station; expanding the historical photovoltaic data based on a data homeomorphic generative adversarial network to obtain a training data set; using the training data set to train a fully connected neural network based on exploratory gradient descent to obtain a feature extraction model; obtaining photovoltaic data to be tested, and inputting the feature extraction model to obtain target feature data; inputting the target feature data into a trained feature dimension reduction model to obtain target dimension reduction data; inputting the target dimension reduction data into a trained classifier model to obtain a load prediction interval. The present application can solve the problem of insufficient training data and improve the generalization ability, stability and model performance of the feature extraction model.