211 results about "Data imbalance" patented technology

Disclosed in the present invention is an improved deep learning model-based refrigeration unit fault detection method. The method uses an LOF algorithm to remove outliers from a fault dataset, and then uses ADASYN technology to solve the problem of data imbalance. In addition, in respect of the problems that existing refrigeration unit fault diagnosis deep learning models are prone to network degradation, and refrigeration unit fault diagnosis models generally lack weighting critical features, the present invention first alleviate, on the basis of ResNet, the problem of network performance degradation which is prone to occur in deep neural network training processes, and then integrates a CBAM for capturing critical features in fault data, so as to improve the feature extraction capability of a network. Experimental results show that the LOF-ADASYN-ResNet-CBAM method provided by the present invention effectively diagnoses refrigeration unit faults.

The invention discloses an industrial part defect sample accurate generation method based on a conditional diffusion model, and belongs to the field of image processing and artificial intelligence. The method forms a closed-loop cooperative system by constructing four deep coupling modules of physical constraint noise scheduling, multi-scale feature coupling, double-domain feedback optimization and adaptive weight adjustment; a defect physical forming mechanism is converted into a dynamic noise scheduling strategy, deep interaction between condition information and a feature map is established at multiple levels of a diffusion network, quality closed-loop optimization is achieved through dual evaluation of a pixel domain and a frequency domain, and training weight is dynamically adjusted according to defect scarcity. And multi-scale accurate control is realized, a quality guarantee closed loop is established, the problem of data imbalance is effectively solved, and the performance of an industrial defect detection model is remarkably improved.

The invention discloses a maritime accident prediction method and device based on interpretable integrated machine learning, and relates to the technical field of maritime affair safety risk analysis, and the method comprises the steps: obtaining accident investigation data, carrying out the preprocessing, balancing the data through a ten-fold layered oversampling method, and carrying out the cross verification training, and determining a performance optimal model by using the test set and carrying out interpretable analysis to explain the influence of the characteristics on the accident prediction result. By constructing a closed-loop'data processing-model optimization-explanation output 'process and adopting SMOTE oversampling and ten-fold layered cross validation training and a heterogeneous base model ensemble learning strategy, the processing capacity of the data imbalance problem of accident categories is improved, the data leakage problem of oversampling is avoided, and the possible bias of a single model is overcome. The interpretability analysis of the model prediction result can quantitatively display the contribution degree of each feature to prediction globally and locally, reveal the nonlinear relationship and interaction effect between the features, and provide transparent interpretation of model decision.

The invention belongs to the technical field of rainfall prediction, and discloses a short and temporary rainfall prediction method based on multi-model random scheduling integration, which develops a robust training and pushing framework based on a continuous rolling prediction strategy, and decomposes long-sequence prediction into manageable stages. According to the method, training is carried out through teacher forcing and planned sampling, error propagation is relieved, and the training process is stabilized. The invention further designs asymmetric encoder-decoders (DSE and AFD) that achieve lower FLOPs than competitive baselines under standardized assessment, where DSE selectively compresses significant features and AFD stepwise reconstructs details to mitigate excessive smoothing problems. Finally, an intensity weighted Gaussian KL divergence loss function is designed, and the key problem of data balance is solved by modeling and predicting on a distribution level and endowing a large weight to a meteorological important heavy rainfall event.

The invention discloses a grouting parameter intelligent feedback method based on rock mass groutability evaluation, which comprises the following steps: collecting geological condition data, rock mass quality data, field test data and grouting process data, and constructing a rock mass groutability grading data set; an SMOTE algorithm is adopted to process data imbalance, input indexes are screened through Pearson correlation analysis, and data are normalized through deviation standardization; an XGBoost classification model is constructed, hyper-parameters are optimized through grid search, the model is trained through K-fold cross validation, and the rock mass groutability grade is evaluated; constructing an XGBoost regression model by taking a rock mass groutability grade, grouting construction data and monitoring data as input, and predicting the maximum value and the minimum value of grouting pressure and slurry density by using an optimal model; and generating a grouting parameter interval according to the predicted maximum value and minimum value, and dynamically feeding back to the grouting equipment to regulate and control parameters. According to the method, scientificity and accuracy of rock mass groutability evaluation and grouting parameter regulation and control can be effectively improved.

The invention discloses an electric energy metering data anomaly detection method based on a generative adversarial network and multi-task optimization, which solves the problem of data imbalance in an electric energy metering data anomaly detection scene by utilizing a dynamic resampling strategy, and trains the generative adversarial network in combination with electric energy metering characteristics. Pseudo samples consistent with real distribution are generated through adversarial training of a generator and a discriminator, and an abnormal sample set is expanded, so that the detection capability of the model on abnormal data is enhanced. By constructing a multi-task learning framework, sharing a feature extraction module and jointly optimizing an anomaly detection task and a load prediction task, the accuracy of anomaly detection and the precision of load prediction are remarkably improved. The method specifically comprises the following steps: a data preprocessing step, a dynamic resampling step, a generative adversarial network training step, a multi-task joint optimization step, and an anomaly detection and load prediction step.

The invention discloses a few-sample target detection method and system based on an aggregation variational prototype. The method comprises the steps of constructing a data set containing a base class and a new class, dividing the data set into a support set and a query set, generating a class prototype by utilizing a P-VAE module in combination with CLIP semantic features and a feature discriminator, realizing bidirectional fusion of the support set and the query set features by means of an MFM module, fusing the query features and the class prototype, and inputting the fused query features and the class prototype into a detection head to complete target detection. The system comprises a data set construction module, a priori variational automatic encoder P-VAE module, a mutual fusion module MFM, a feature fusion module and a target detection module. According to the scheme, by introducing semantic priori, optimizing prototype generation and feature interaction, the problems of data imbalance and insufficient new class feature representation in a few-sample scene are solved, improvement of new class detection precision is verified on PASCAL VOC, MS COCO and other data sets, and an effective solution is provided for target detection of sample scarce scenes such as medical images and rare species monitoring.

The invention relates to the technical field of data processing, and discloses a defect identification method and device for defect data, equipment and a medium, and the method comprises the steps: dynamically adjusting the attention weight of a defect identification model for difficult-to-classify samples based on the feature distribution and inter-class relationship of each sample in a training sample set; extracting text features and image features of the input data through a multi-mode encoder, and generating a fusion control signal; based on the fusion control signal and the attention weight, generating synthetic defect data consistent with target defect semantics through a controllable diffusion model; mixing the synthesized defect data with an original training sample set, and training a defect recognition model after a balanced data set is constructed; and carrying out defect classification and positioning on an input image by utilizing the trained defect identification model, and outputting defect category and position information. According to the method, the problems of data imbalance and sample scarcity can be effectively solved, and the accuracy and robustness of defect identification on the defect data are remarkably improved.

The invention discloses a large model-based production breaking data analysis method, equipment and a medium. The method comprises the following steps: acquiring multi-channel production breaking data through a distributed data acquisition system, and realizing automatic classified storage of the data by adopting OCR and text extraction technologies; a deep learning model is utilized to extract a production breaking field entity, and an enterprise production breaking knowledge graph is constructed through a mutual attention mechanism; an oversampling algorithm is applied to process a data imbalance problem, and a multi-dimensional feature system is constructed in combination with a knowledge graph; a Qwen-plus large language model and an improved V-detector algorithm are adopted to realize dynamic risk early warning; similar historical cases are retrieved based on the knowledge graph, and a personalized disposal scheme is generated in combination with rule reasoning; according to the method, the knowledge graph and the deep learning technology are creatively fused, the dynamic monitoring, accurate early warning and intelligent decision support of the production breaking risk are realized, and the accuracy and timeliness of the production breaking data analysis are remarkably improved. The method can be widely applied to the field of financial institution, court and enterprise risk management.

The invention discloses a voice and music collaborative generation method and system based on a dynamic mixed attention and expert architecture, terminal equipment and a medium, and relates to the technical field of audio generation. The method comprises the following steps: acquiring multi-modal input containing at least one of audio, text and vision, performing embedding processing on the multi-modal input, and mapping the multi-modal input to a unified space to obtain a fusion sequence; setting a plurality of attention heads, dynamically selecting the attention heads to activate experts through attention head routing gating, and weighting and aggregating calculation results to obtain mixed attention features; generating expert probability distribution through expert set routing gating, dynamically selecting an expert set to activate experts, and calculating task adaptation features; and performing audio generation processing based on the audio language modeling head to obtain a voice or music result. According to the method, collaborative generation of voice and music, dynamic allocation of computing resources, balance of domain-specific and cross-domain general knowledge learning are realized, the problems of task conflict and data imbalance are solved, and the audio generation quality and efficiency are improved.

The invention belongs to the technical field of visual anomaly detection, and discloses an image small sample anomaly detection system and method based on computer vision. The system is composed of a cloud-edge environment perception federated learning module, an edge end training and pushing integrated lightweight detection module, a small sample adaptive loss optimization module and an edge-cloud collaborative iteration updating module, and a category balance comparison loss function is designed through the small sample adaptive loss optimization module. In combination with inverse frequency weighting, dynamic threshold adjustment and environmental context constraints, the model focuses on rare samples, sample differences are accurately measured, and abnormity is accurately judged according to the environmental background; self-supervised learning is utilized to generate a positive sample pseudo label, data enhancement is performed on a negative sample, the data quantity and diversity are expanded, the small sample utilization rate is improved, the influence of data imbalance on model training is effectively relieved, the labeling cost and subjective errors are reduced, the data quality is optimized, the generalization ability of the model to an abnormal mode is enhanced, and the detection effect is improved.

The invention discloses a semi-supervised prognosis prediction system based on irregular sampling medical data pre-training, and the system comprises a clinical electronic medical record data collection and preprocessing module which automatically collects original clinical electronic medical record EHR data from a medical database; and the pre-training module is used for receiving the patient feature vector sequence output by the clinical electronic medical record data acquisition and preprocessing module and carrying out feature representation learning on the preprocessed clinical time sequence data by utilizing a combined multi-task self-supervised learning mechanism. And the classifier fine tuning and pseudo-label iterative optimization module is used for training the pre-trained model through fine tuning of samples with labels and carrying out iterative optimization through guidance of pseudo-labels to obtain a prediction result. And the application display module is used for displaying and outputting a post-hospital-admission vital sign sequence and a prediction result. According to the method, irregular sampling and missing data are effectively processed, information loss is avoided, and the prediction capability of the model and the performance of the model under the conditions of data imbalance and label scarcity are improved.

The invention relates to the technical field of computers, and discloses a subject classification model construction method and system, electronic equipment and a storage medium, and the method comprises the steps: obtaining academic paper bibliography data based on an academic database, and constructing an initial training data set; identifying minority category subjects of which the sample quantity is lower than a preset threshold value, generating synthetic data containing chapters and keywords, and labeling corresponding subject categories; mixing the synthetic data with real data in the initial training data set, and constructing a balanced mixed training data set; splicing a text based on the chapter and the keyword of each sample in the mixed training data set, and generating a multi-level fusion feature; and taking the multi-level fusion features as input, accessing a full-connection classification layer to construct a model, carrying out end-to-end training based on a mixed training data set, and adjusting and optimizing model hyper-parameters to obtain a final subject classification model. According to the method, data imbalance can be effectively relieved, existing labeling resources are fully utilized, and the model generalization ability is improved.

The invention relates to an unbalanced bearing data fault diagnosis method based on an improved condition Wasserstein generative adversarial network. The problem of bearing data imbalance under a small sample condition is effectively relieved. According to the technical scheme, the method comprises the following steps: firstly, processing a bearing vibration signal by adopting a sliding window overlapping sampling strategy, and converting the bearing vibration signal into a time-frequency image by applying continuous wavelet transform so as to enhance fault feature representation; secondly, an improved condition Wasserstein generative adversarial network is used for carrying out data enhancement, the network measures sample distribution difference through a Wasserstein distance, introduces gradient penalty and L1 loss, combines spectrum normalization and a self-attention mechanism, stabilizes a training process and prevents gradient explosion, and meanwhile, generates a specific category of high-quality samples through condition information; and finally, carrying out accurate diagnosis by adopting a convolutional neural network equipped with a global attention mechanism. According to the method, the common collapse and instability problems of the generative adversarial network can be avoided, and the precise diagnosis of the rolling bearing can still be realized under the conditions of small samples and unbalanced data.

The invention belongs to the field of natural language processing in computer intelligent information processing, and discloses a document level relation extraction method and system based on information gain and prototype comparative learning. The invention provides a document-level relation extraction model based on a graph structure, which considers two aspects of extracting more accurate node features and relieving data imbalance. The problem that an existing document-level relation extraction model generally adopts a graph-based model and faces inherent data imbalance is solved. At present, the problems that noise interference is caused by irrelevant nodes and edges in the node feature updating process, the learning ability of a model to a real relation is insufficient due to too many negative samples in a document, and all different relation types cannot be accurately predicted through multi-label classification exist in research.

The invention discloses a point cloud single tree segmentation method and system. The method comprises the steps of obtaining point cloud data of a target tree and an environment in a power transmission corridor area; constructing a segmentation network model, and performing model training through the optimized loss function and adjustment of the adaptive learning rate to obtain an optimized segmentation network model; inputting the preprocessed point cloud data into the optimized segmentation network model, calculating the category of each point through an activation function, and outputting the category prediction of each point; and point cloud segmentation is carried out based on the category, and a segmentation result is evaluated. According to the invention, point cloud data segmentation is carried out through the segmentation network architecture, the learning rate and the loss function in the architecture are optimized, and the recognition and segmentation precision of the tree monomers is improved in combination with the probability distribution of deep learning; the robustness in a complex environment is enhanced, the problems of noise, sparse point clouds and data imbalance can be effectively solved, and the method is suitable for tree monitoring, accurate positioning and safety evaluation of the transmission line corridor.

The invention discloses a thin-wall shell plastic part injection molding process parameter multi-objective optimization method, which is based on an RIME-RF-MOGWO framework, takes a simulation sample as a research object, selects a volume shrinkage rate and a buckling deformation amount as optimization objectives, and firstly adopts SMOTE to process a data imbalance problem; establishing a nonlinear mapping relation between the process parameters and the quality target by using RF; an RIME is introduced to optimize the hyper-parameter of the RF; multi-objective optimization of process parameters is realized in combination with MOGWO, and a Pareto frontier solution set is obtained through non-dominated sorting and a congestion degree control mechanism. A multi-round optimization and simulation verification result shows that the multi-objective optimization method for the injection molding process parameters of the thin-wall shell plastic part can effectively obtain an optimal process parameter combination, the volume shrinkage rate is reduced by 19.02%, the buckling deformation amount is reduced by 50.63%, and the molding quality of the thin-wall plastic part can be remarkably improved.

The invention discloses a multi-fault diagnosis method for lithium battery sample scarcity and data imbalance, and the method comprises the steps: firstly constructing a feature extraction model based on a residual neural network, introducing an improved multi-factor imbalance index (MFI), carrying out the analysis of real-time monitoring batch feature distribution through employing a minimum spanning tree, and carrying out the real-time monitoring of the real-time monitoring batch feature distribution; therefore, the loss function and the sample weight are dynamically adjusted, and the learning stability of majority classes and the recognition precision of minority classes are both considered. On the basis, a prototype vector of a normal working condition is obtained through sample feature mean value calculation, and a prototype network (ProtoNet) is constructed to serve as an anomaly detector; after features of a test sample are extracted through the ResNet-MFII module, the Euclidean distance between the test sample and a normal prototype is calculated, if the Euclidean distance exceeds a set threshold value, it is judged that the test sample is abnormal, and detection of unknown or rare faults is achieved. The system finally outputs fault types and abnormal alarms, and high-precision recognition of multiple types of faults such as short circuit and aging of the lithium ion battery is achieved.

The invention discloses a method and a system for detecting a complex multi-step attack based on an interpretable graph neural network, and a storage medium. The method comprises the following steps: generating a strong negative sample by using priori attack knowledge; inputting the overall graph into a detection model to obtain an attack detection result; and inputting the attack event and the detected attack behavior into an interpreter to obtain the interpretation of the attack behavior. Constructing a traceability graph and an attack mode graph from a system log, aligning the generated traceability graph and attack mode graph, and enhancing an overall graph; pre-training an encoder of a multi-step attack detector on the continuous time dynamic heterogeneous graph by using comparative learning; finely adjusting the model on a small number of real attack samples based on a pre-training model; the relevance between an abnormal event and a preorder event is calculated by using a sniffer, and then an interpretable sub-graph is searched and output by using a Monte Carlo tree under the guidance of the sniffer by a digger. The problems that an existing method is difficult to solve the problem of data imbalance, low in interpretability and the like are solved.

The application discloses a method for generating electrocardiogram based on a diffusion model and synthesizing a customizable cardiac cycle, relates to an electrocardiogram generation method, and aims at solving the problems of the existing electrocardiogram generation method, such as data imbalance, poor privacy protection and the incapability of generating specific pathological signals. The application takes electrocardiogram semantic labels as conditional input, simultaneously inputs noise and diffusion time steps into a deep generation model, and generates electrocardiogram; the deep generation model takes a diffusion model as an overall architecture, introduces a converter model to learn long-term dependencies in electrocardiogram signals, and simultaneously introduces a semantic electrocardiogram batch normalization module to accurately learn local ECG semantic features. The electrocardiogram signal generated by the application can accurately follow the provided electrocardiogram semantic information, customize electrocardiogram with real physiological significance, and improve the data imbalance problem and the privacy protection.

The invention discloses a metal surface multi-scale defect recognition and classification method, system and device and a medium, and relates to the field of metal detection.The method comprises the steps that metal surface images are collected and preprocessed, and a preprocessed image set is obtained; performing data enhancement on the preprocessed image set by adopting a multi-layer strategy search and gradient matching mechanism to obtain an enhanced image set; based on the enhanced image set, adopting an SPGP-YOLOv11 network to generate a candidate detection frame of the metal defect; based on the candidate detection frame, determining the type of the metal defect by adopting a pre-trained classifier and a pre-trained two-state reinforcement learning algorithm; the classifier and the two-state reinforcement learning algorithm are cooperatively trained. The technical bottlenecks of the existing method in the aspects of multi-scale defect detection precision, small target detection effect, model real-time performance, data imbalance, insufficient robustness and the like can be solved.

The invention provides a threat detection method and device, and the method comprises the steps: constructing a time sequence diagram sequence corresponding to each time window through the preprocessed multi-source log data of a target network; embedding nodes in the time sequence diagram sequence corresponding to each time window by using a relational graph neural network to obtain an embedded sequence matrix; inputting the embedded sequence matrix into a Transform model, and performing autoregressive prediction to obtain a node prediction matrix of a next sub-graph node; constructing an attention score matrix through the predicted node prediction matrix of the next sub-graph node, so as to predict a connection matrix at the next moment; and calculating an error between the connection matrix at the next moment and the real adjacent matrix at the next moment to determine whether a security threat exists at the next moment. According to the scheme, the technical problems of data imbalance and data annotation in the prior art are solved, and efficient identification of low-frequency and hidden attack behaviors is achieved.

The invention discloses an aero-engine vibration signal generation method under extreme imbalance, and belongs to the technical field of aero-engine fault signal generation. The method comprises the following steps: acquiring an original vibration signal at a target position of the aero-engine, and performing preprocessing operation; taking the preprocessed real vibration signal as a training sample, constructing a forward noise adding process, and training the improved one-dimensional diffusion generation network to learn a reverse denoising mapping relation; and inputting random Gaussian noise into the one-dimensional diffusion generation network after training convergence, and generating a target vibration signal matched with a real vibration signal feature through a reverse diffusion denoising process. Through targeted improvement of the diffusion generation network, the problem of data imbalance is effectively relieved, the model deployment cost is reduced, the signal generation capability of the model for minority types of faults is improved, and the method is suitable for fault signal generation scenes of aero-engines under complex working conditions.

The invention discloses a real-time evaluation method for reliability of marine autonomous surface ship mechanical equipment, which comprises the following steps of: S1, constructing a synthetic data set of operation of the ship mechanical equipment according with an actual degradation rule, and solving the problem of imbalance between small samples and data; s2, evaluating a health index value of the ship mechanical equipment by adopting a comprehensive weighting method and a plurality of types of processed operation characteristic data; s3, constructing an unsupervised health index prediction model for predicting a health index value, and training the unsupervised health index prediction model to obtain a trained unsupervised health index prediction model; s4, performing Weibull distribution fitting on the health index value predicted by the trained unsupervised health index prediction model to obtain a shape parameter and a scale parameter of the ship mechanical equipment under Weibull distribution; and S5, real-time evaluation of the reliability of the ship mechanical equipment is realized based on the shape parameters and the scale parameters, early fault detection is supported, and the safety of the MASS is improved.

The invention provides a sodium ion battery safety early warning method and device based on multi-dimensional feature fusion and a neural network, and a storage medium, and belongs to the technical field of sodium ion battery safety management. The problems of low anomaly detection sensitivity, poor aging adaptability, high false alarm rate and the like in a traditional safety early warning method are solved; according to the method, a TCN-Bi-GRU hybrid neural network is constructed, millisecond-level voltage step characteristics of sodium ions are accurately captured through time sequence convolution, and a long-time temperature rise trend is analyzed in combination with a bidirectional gating unit; aiming at a specific capacity attenuation mode of the sodium battery, designing a three-level dynamic threshold collaborative optimization system based on SOH compensation, and realizing self-adaptive adjustment of an early warning boundary through a health state compensation function and an intelligent learning algorithm; a self-adaptive sample enhancement strategy and a multi-stage training method which are combined with sodium electrochemical characteristics are developed, safety early warning model deviation caused by data imbalance is effectively relieved, and the method can be applied to safety early warning of the sodium ion battery.

The invention discloses a power system operation risk key feature extraction method, system and device and a medium, and the method comprises the steps: giving a historical random variable, and carrying out the first processing, and obtaining low-dimensional clustering data; constructing a conventional graphic vine model, determining optimal parameters of the conventional graphic vine model based on the low-dimensional clustering data, and generating a sampling sample set; performing simulation before and after a fault based on the sampling sample set, and extracting features and labels of the training set and the test set by using an encoder; and constructing a power system safety rule according to the extracted features, and performing performance evaluation on the power system safety rule. According to the method, the R-vine Copula model and the depth automatic encoder are combined, so that the accuracy and robustness of risk assessment of the power system are remarkably improved. The method not only solves the problems of data imbalance and complex dependence caused by high-proportion new energy access, but also is superior to a traditional method in key indexes such as precision and F1 score, and provides more reliable decision support for safe operation of a power system.

The invention relates to the technical field of wind power generation, and discloses a wind turbine generator fault diagnosis data processing method based on a machine learning algorithm, which adopts multi-source data cleaning and working condition self-adaptive slicing to ensure data quality. Automatic discovery and explanation of fault subclasses are realized through multi-algorithm consensus clustering and physical mechanism mapping; an unbalance index comprehensively considering quantity and classification difficulty is innovatively proposed, and a data set is accurately balanced in combination with a physical constraint oversampling technology; and finally, through hierarchical training and dynamic weight optimization, a high-precision and high-robustness interpretable diagnosis model is constructed. According to the method, high-precision traps caused by data imbalance of the model are effectively avoided, the recognition capability of early weak faults is remarkably improved, reliable technical support is provided for predictive maintenance of the wind turbine generator, and the operation and maintenance cost and the fault risk are greatly reduced.

The invention discloses a medical data intelligent identification method and apparatus. The method comprises the steps of obtaining to-be-identified target data; inputting the target data into a pre-trained intelligent recognition model to obtain a recognition result output by the intelligent recognition model, the recognition result at least comprising all sample categories of the target data and statistics of the number of the categories; wherein the intelligent identification model is obtained by training and optimizing a pre-constructed neural network by using a training data set, the training data set trains the neural network to obtain an initial network, and the initial model is optimized by using a preset optimization target to obtain the intelligent identification model. According to the intelligent recognition model based on feature distribution robustness optimization, by optimizing feature distribution, automatically exploring and constructing an optimal and most robust feature space and introducing anti-noise classification loss, the problems of medical data imbalance, annotation noise and generalization in the prior art are solved.

The present application relates to the technical field of data processing, and more particularly to a data privacy protection and compliance use method and system, the present application is provided with data acquisition module, terminal pretreatment module, terminal analysis module and terminal identification module, the access offset tendency parameter of the access terminal is determined through the terminal pretreatment module, the feature access terminal is screened, the data access risk curve is constructed for a single feature access terminal based on the data imbalance tendency parameter through the terminal analysis module, whether the data isolation is carried out to the feature access terminal based on the risk representation coefficient of the feature access terminal through the terminal identification module, and then, the access terminal with abnormal risk is quickly identified according to the actual access characteristics of the access terminal, the data processing mode of the access terminal with abnormal risk is adaptively adjusted, and the processing efficiency and reliability of the data privacy protection and compliance use system are improved.

The application discloses a train axle temperature anomaly recognition method based on a generative adversarial network and ensemble learning, which comprises the following steps: collecting operation data in actual operation of an urban rail train, obtaining a two-class data set with unbalanced categories after preprocessing, and dividing the data set into a training set and a test set; training a constructor and a discriminator of the generative adversarial network by using abnormal data samples in the training set, and realizing automatic network parameter adjustment by using a Bayesian optimization algorithm; synthesizing abnormal samples by using the trained generative adversarial network model, and jointly constructing a training set with balanced categories with the original training set; filtering and screening noise samples by using a cross-committee filtering technology; constructing an axle temperature anomaly recognition classifier by using an AdaBoost method, training the ensemble learning model by using the training set, and inputting the test set to obtain a test result. The application solves the problems of missing of the axle temperature abnormal samples of the urban rail train and data imbalance, and improves the accuracy and correctness of the axle temperature anomaly recognition.