65results about How to "Reduce imbalance" patented technology

The invention discloses a balance control scheduling method and system based on distributed new energy, and the method comprises the steps: obtaining power grid parameters, and constructing a time-space characteristic model of multi-type adjustable resources and a distributed scheduling frame based on the obtained parameters; based on the space-time characteristic model and the distributed scheduling framework, establishing a distributed new energy participation power grid adjustable and controllable resource group internal cooperation strategy; establishing a distributed new energy participation power grid dispatching collaborative interaction and balance optimization control strategy according to the collaborative strategy; and realizing active / reactive coupling balance collaborative optimization of the high-permeability distributed new energy access power grid based on a collaborative strategy and an optimization control strategy. According to the invention, the utilization rate of new energy can be improved, cooperative work of multiple types of adjustable and controllable resources is realized, the flexibility and stability of a power grid are improved, dynamic adjustment of a power system is realized, and optimal matching of power supply and demand at different time and space scales is ensured.

The utility model relates to a crutch with adjustable height and angle, which comprises a crutch body and a stabilizing component, the stabilizing component is arranged in the crutch body, and the stabilizing component is used for stabilizing the crutch body; and the stabilizing assembly comprises a round cover, a round groove is formed in the round cover, a corrugated sleeve is fixedly connected to the bottom of the round cover, and a bottom plate is fixedly connected to the bottom of the corrugated sleeve. When the bottom of the walking stick body is in contact with the inclined ground, the bottom of the walking stick body can be always parallel to the ground under the running fit of the round ball and the round groove, and even if the walking stick is on the uneven or inclined ground, good contact can be kept, and more reliable support can be provided; according to the walking aid, shaking and unbalance caused by uneven ground when a user walks can be effectively reduced, meanwhile, the reset spring and the corrugated sleeve can provide certain buffering and damping effects in the using process, and the hand pressure of the user is relieved.

The invention relates to the technical field of computer machine vision, and discloses a plant growth condition monitoring method, system and device and a storage medium, and the method comprises the steps: obtaining a target plant image, inputting the preprocessed target plant image into an improved YOLOv9 model (comprising a backbone network, a feature fusion network and a prediction classification network), and obtaining a target plant image; obtaining a target plant detection image; performing feature extraction on the preprocessed target plant image through a backbone network to obtain plant multi-scale features; performing feature conversion, branch processing and feature fusion on the plant multi-scale features through a feature fusion network to obtain plant fusion features; performing classification processing on the plant fusion features through a prediction classification network to obtain a target plant detection image; and inputting the target plant detection image into an image semantic segmentation model (including an image super-resolution network and an image segmentation network) to obtain a plant binary image. The method effectively improves the accuracy of plant growth condition monitoring.

The invention discloses a semi-supervised multi-class medical image semantic segmentation method based on prototype comparative learning. The method comprises the following steps: dividing a marked image data set and an unmarked image data set into a training sample set and a test sample set; constructing a deep learning model of semi-supervised semantic segmentation, constructing a prototype-like feature library of a main and branch network, constructing a total loss function of the deep learning model, inputting the training sample set into the deep learning model for training to obtain a semantic segmentation model, wherein a data enhancement scheme of random rotation and random flipping is applied in each round of training; and testing the trained deep learning model by using the test sample, and outputting a test sample image segmentation result. According to the method, the workload that a doctor needs to label the data in a daily medical image processing task is reduced, and the problem that data labeling is scarce generally existing in the field of medical image processing is solved.

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 relates to the technical field of circuit breaker control, and discloses a high-voltage circuit breaker intelligent control system based on a positive and negative zero sequence control algorithm, and the system comprises a cascaded H-bridge static synchronous compensator which is in star connection, and each phase of the cascaded H-bridge static synchronous compensator is composed of a plurality of H-bridge units; the data acquisition module is used for acquiring voltage and current signals of the three-phase system in real time; the control module is used for operating a positive and negative zero sequence control algorithm and realizing voltage fluctuation optimization and unbalanced load compensation through independent control of three-sequence components; the communication module is used for data interaction between the control module and an upper computer; and the connecting inductor is used for connecting the static synchronous compensator and a power grid so as to reduce harmonic interference and provide necessary impedance. By independently controlling the positive sequence component, the negative sequence component and the zero sequence component, the technical effect of reducing the unbalance degree of the power grid is achieved, the problems of response lag and high unbalance degree under the condition of rapid load change in the traditional technology are solved, and the electric energy quality of the system is improved.

The invention discloses a multi-source data fusion-based air volume dynamic adjustment method of a fume hood for a clean environment, particularly relates to the technical field of clean environment control, and aims at solving the problem that air volume adjustment of an existing fume hood cannot give consideration to front air suction safety, pressure difference stability and overall energy conservation. A basic operation data set is formed by collecting and filtering window opening and cabinet air speed, an air volume demand value and an initial priority parameter are calculated according to the data set and a room use state mark to form a demand data set, and a deviation memory parameter and a fluctuation influence parameter are calculated by fusing room pressure difference data and the availability of an exhaust fan. Obtaining a stable control priority adjustment coefficient through a time decay score learning model, correcting the priority, distributing the target exhaust air rate and the exhaust fan output, forming a target distribution result, comparing the target distribution result with the actual air valve opening degree and the exhaust fan output state, calculating the adjustment amount, and issuing a control instruction according to the priority; and the deviation record feedback updating coefficient and the demand value are verified in the control period to realize self-adaptive adjustment.

The invention discloses a few-sample fault diagnosis method based on TFCD fusing time sequence characteristics, and the method comprises the steps: inputting a preprocessing state signal into a trained fault diagnosis model, so as to enable the fault diagnosis model to carry out the fault diagnosis of a core part. The fault diagnosis model is obtained by training a plurality of training samples generated by a trained TFCD model, and the TFCD model extracts discriminative time sequence features by introducing a long short term memory network, and embeds the discriminative time sequence features as static condition vectors into a DDPM model. According to the method, the model can effectively capture the long-range dependency relationship and the dynamic mode characteristics in the vibration signal, so that the generated sample keeps good coherence in the time domain, meanwhile, the specific periodic impact mode of the fault is reserved, and the physical rationality of the generated sample is remarkably improved. Therefore, accurate diagnosis under cross-working conditions can be realized.

The present application belongs to the technical field of installation detection of clothes processing equipment, and discloses an installation detection method, an installation detection device and clothes processing equipment. The installation detection method comprises the following steps: after the installation of the clothes processing equipment is completed, a detection operation process of the clothes processing equipment is performed; during the detection operation process of the clothes processing equipment, the balance state of the clothes processing equipment is judged according to the pressure values of a plurality of detection positions at the bottom of the clothes processing equipment, and balance adjustment prompt information is reported; and installation prompt information is reported according to the noise value and the vibration acceleration value of the clothes processing equipment. The installation detection method can detect the clothes processing equipment after the installation is completed, analyze the causes of the eccentricity problem of the equipment, and facilitate quick adjustment.

The invention discloses a lactobacillus delbrueckii subsp. Bulgaricus CCFM1520 for transforming curcumin to produce tetrahydrocurcumin to relieve nervous anxiety, and belongs to the technical field of microorganisms. The lactobacillus delbrueckii subsp. Bulgaricus CCFM1520 screened by the invention has the capability of converting curcumin, and can utilize curcumin and generate tetrahydrocurcumin. The invention further provides a synbiotic preparation prepared from the lactobacillus gasseri CCFM1481 and curcumin, the synbiotic preparation can be used as an anti-fatigue functional medicine, the individual behavioral memory level can be improved, the imbalance of body neurotransmitters (DA, 5-HT and NE) can be slowed down, the expression of body neuroprotection related genes (BDNF, TrkB, DRD1 and TH) can be improved, and the synbiotic preparation has a huge application prospect.

The present invention discloses a rolling bearing fault diagnosis method based on L-ICGAN, which belongs to the field of fault diagnosis technology, and includes the following steps: step S1, data preprocessing, collecting vibration acceleration signals under different states, converting the one-dimensional time domain signal into a two-dimensional logarithmic spectrum diagram through logarithmic spectrum transformation, and dividing the preprocessed data into a training set and a test set; step S2, network training, initializing the L-ICGAN model through the training set, and training the L-ICGAN model; step S3, fault diagnosis, introducing the divided test set into the trained L-ICGAN model. The present invention adopts the above-mentioned rolling bearing fault diagnosis method based on L-ICGAN, which can not only effectively remove redundant information and show stronger frequency characteristics, but also increase the generation probability of minority category samples and alleviate the problem of data imbalance.

The invention belongs to the technical field of peanut planting, and particularly relates to a peanut strip rotary tillage stubble cleaning seeder and a seeding method.The peanut strip rotary tillage stubble cleaning seeder is characterized in that one end of a seeding frame is fixedly connected with a fixing frame, a transmission shaft set is arranged on the inner side of the fixing frame, and the end, away from the fixing frame, of the seeding frame is rotationally connected with a plow column; a plurality of soil ploughing blades are fixedly connected to the annular side of the plow column, a stirring and dispersing column is rotationally connected to the center of the seeding frame and close to one end of the plow column, a plurality of linkage columns are fixedly connected to the annular side of the stirring and dispersing column, and the transmission end of the transmission shaft set is in meshed connection with the linkage columns. Soil scattered by the ploughing blade is further vibrated and scattered through the stubble cleaning assembly, rhizomes of crops in the previous batch in the soil are picked out, the cultivated soil is finer, the number of the rhizomes in cultivated land is reduced, decomposition and consumption of nutrients in the soil by the residual rhizomes are reduced, and unbalance of the soil nutrients is reduced.

The invention relates to the field of laser radar signal processing, in particular to a laser radar cloud layer detection algorithm based on deep learning, a storage medium and an electronic device. The laser radar cloud layer detection algorithm based on deep learning comprises the following steps: preprocessing a laser radar backscattering signal; establishing a laser radar cloud layer detection data set; building a laser radar cloud layer detection model; training the laser radar cloud layer detection model and storing a training result; the laser radar signals collected in real time are preprocessed; and inputting the real-time preprocessing signal into the trained laser radar cloud layer detection model for cloud layer detection. The cloud layer detection method has the advantages that efficient and accurate cloud layer detection can be achieved by using the deep learning model, the cloud layer detection method can adapt to environment changes under different meteorological conditions, and missing detection and false detection are effectively reduced.

The invention relates to the technical field of remote sensing image fine-grained target detection, in particular to a two-way guided remote sensing image fine-grained target detection method. Comprising the steps of 1, acquiring image data, and dividing the image data into a training set and a test set; 2, constructing a target detection network, and obtaining image multi-scale features from the preprocessed image through a feature extraction network; then, a coarse-grained detector screens out a target candidate area and judges whether the target candidate area is a foreground or a background; then, a fine-grained detector carries out finer classification and positioning on the candidate areas; and step 3, carrying out performance evaluation on the test set so as to test the detection capability of the model on the fine-grained target in an actual scene. According to the method, for the problems of class sample imbalance and easy confusion of fine-grained classes in fine-grained target detection, the remote-sensing image fine-grained target detection capability is effectively improved by constructing a layered detection network architecture and combining an instantaneous confidence coefficient signal and class accumulation learning quality.

The application relates to a video instance segmentation method based on a dynamic convolution decomposition of a lightweight attention mechanism; the method first inputs a video frame, and a backbone network independently extracts a feature map of each frame in the video; then the feature map extracted by the backbone network enters an HQT encoding and decoding module, the position change of an instance in each frame is accurately positioned by combining an output head, and three prediction branches are used to supervise model training; in the application, a convolution layer of an encoding network adopts dynamic convolution decomposition, a more compact model is obtained, model training is easier, the required parameter quantity is greatly reduced, the training speed is improved, and the training time is shortened; a lightweight HQT encoding and decoding module is provided, which helps to reduce model parameters and improve efficiency; the loss function is improved, model training is more stable, the convergence speed and convergence precision are improved, and the problems of imbalance between foreground and background in samples and imbalance between foreground categories under a long tail condition are relieved.

The present invention belongs to the field of power systems, and specifically relates to a power grid reliability assessment method based on fine-grained data drive; the method includes randomly extracting power grid system status data using the Monte Carlo method; calculating the power difference of the system DC power flow before and after the power grid system is disconnected; inputting the power difference and other power data into the power flow module as the input feature vector of the sample to be tested; calculating the regression result and classification result of the sample to be tested, and determining the problem sample; inputting the input feature vector of the problem sample into the optimal power flow module, calculating the classification result of the problem sample, and determining the fault sample; performing minimum load shedding analysis on the fault sample, and according to the analysis result, combining the variance coefficient to calculate the uncertainty of the fault sample, and outputting the reliability assessment result of the fault sample. The present invention can be widely used in the online calculation of the operation reliability of large power grids, and is particularly suitable for the scene after large-scale access of high-proportion new energy.

The invention discloses a hybrid intrusion detection method suitable for incomplete data of the Internet of Things. The method comprises the following steps: acquiring an open source data set and preprocessing the open source data set; configuration of a training environment; constructing and training a multi-classification-stage algorithm model, wherein the multi-classification-stage algorithm model is used for carrying out weighted aggregation on four integrated learning algorithms, namely an RF algorithm, a category type feature lifting algorithm CatBoost, an XGBoost algorithm and a lightweight gradient lifting machine algorithm LightGBM; a binary classification stage algorithm model is constructed and trained, according to the binary classification stage algorithm model, weighted summation is carried out on output abnormal scores of an auto-encoder, a single-class support vector machine and an isolated forest algorithm to obtain an aggregation abnormal score, and the aggregation abnormal score is compared with a threshold value to obtain a final detection result. The method can better adapt to a data set with incomplete classes, further provides the self-adaptability of the system, and can effectively aggregate the training results of various algorithms.

The invention belongs to the technical field of power distribution network leakage current detection, and particularly relates to a CNN and FiLM-based leakage current type intelligent identification method. Comprising the following steps: S1, collecting real-time operation data of a typical power supply area, obtaining leakage current waveforms and related environment characteristic parameters under different working conditions, and constructing an original leakage current sample data set; s2, a CNN-based leakage current classification model is constructed, a FiLM condition modulation module is introduced, a multi-task learning framework is used at the tail of the network, a main task is leakage current type identification, and an auxiliary task is grounding system discrimination; s3, using a weighted cross entropy loss function to alleviate a class sample imbalance problem; an OneCycleLR dynamic learning rate scheduling strategy is introduced; s4, evaluating the performance of the model on the test set, and using the accuracy and the confusion matrix as evaluation indexes; according to the method, high-precision identification of multiple types of faults such as single-phase grounding, arc type electric leakage and direct current system electric leakage is realized, different grounding systems can be adapted, and the accuracy and robustness of system diagnosis are improved.

The invention provides a wind-light-fire-storage integrated collaborative planning method and system, and relates to the field of wind-light-fire-storage integration, and the method comprises the steps: generating a first quantification result according to the site selection related information of a to-be-planned region and the influence of regional quantification meteorological factors on the wind-light output of the to-be-planned region, determining a resource evaluation result of the to-be-planned region, and carrying out the resource evaluation of the to-be-planned region. Determining a target position; according to the building related information of the load area, quantifying the load influence of the meteorological factors on the load area, and generating a second quantification result; predicting power load information of the load area according to the second quantification result; the method comprises the steps of generating an initial planning scheme, performing source load unbalance degree evaluation on the initial planning scheme according to power load information of a load area and historical meteorological information of a target position and a first time scale, generating a source load unbalance degree evaluation result, and generating a target planning scheme according to the source load unbalance degree evaluation result. And the resource utilization rate and the power supply reliability of wind-solar-thermal storage integration are improved.

The utility model discloses a mini robot dog structure, including main part and four leg parts, the leg part rotation is arranged on the main part, be provided with four drive mechanisms in the main part, four drive mechanisms each with a leg part is connected, four leg parts can rotate respectively, and the flexibility is higher, the leg part includes front casing, rear casing and soft rubber antiskid part, and the front casing is connected through screw with rear casing, and the soft rubber antiskid part is arc, can promote the continuity of movement and enhance the adaptability of robot dog in complex terrain, the outer surface of soft rubber antiskid part is provided with a plurality of antiskid grooves laterally, can enhance the antiskid effect, and the front and rear sides of soft rubber antiskid part extend outward and form the clamping part, the bottom of front casing and rear casing is provided with the installation port matched with soft rubber antiskid part, the inboard of installation port is provided with the clamping groove, and the clamping part is placed into the clamping groove, and can be replaced individually after the wear failure of soft rubber antiskid part, to reduce the maintenance cost.

The present application relates to the field of industrial data processing, and provides an industrial data virtual sample generation method based on physical constraints and attention mechanisms, which constructs feature data in response to an industrial data virtual sample generation task; the feature data is input into a virtual sample generation model for training, and corresponding candidate virtual samples are generated after model training, the virtual sample generation model includes a classification encoder, a mean-variance encoder, a Gaussian mixture latent space and a decoder, the candidate virtual samples are sequentially subjected to category validity screening, physical constraint compliance screening and distribution similarity screening to construct a high-quality candidate sample pool, and a diversity saturation critical point is determined based on feature entropy change trend analysis, and an optimal virtual sample set is output. Through the collaborative fusion of the attention mechanism, Gaussian mixture latent modeling and physical constraint loss, combined with multi-dimensional screening and entropy change monitoring mechanism, unified control of virtual sample generation in a small sample industrial scene is realized.

A data re-labeling classification algorithm based on clustering learning comprises the following steps: after preprocessing the expression image data in the original sample, a feature extraction module is used to extract a feature vector to form sample data; a dynamic clustering module splits a sample data set of a minority class into several sub-class data sets, and a sample data set of a majority class remains unchanged; the extracted feature vector is used as the input of the dynamic clustering module; if the sample data belongs to the sample data set of the minority class, the sample data is assigned to the nearest sub-class data set according to the distance between the sample data and the clustering center, and then a pseudo label is assigned to all sample data; the sub-classes are trained by using the pseudo label; and a label mapping module maps the prediction result of the sub-class data set to which the sample data belongs back to the real label space based on the mapping relationship learned in the training process. The application can improve the classification and recognition performance of an unbalanced expression data set, and fully excavates and utilizes the potential information of the sample data set.

According to the network intrusion detection enhancement method based on the diffusion model provided by the invention, noise is gradually added to training data through the diffusion model, and then a denoising process is learned to model data distribution. Based on this, the method can enhance the training sample by synthesizing the virtual sample, especially the category with a small number of enhanced samples, thereby solving the problem of low detection precision caused by sample imbalance. Meanwhile, the method can enhance the adversarial samples, and introduces a certain degree of change to synthesize more diversified adversarial samples while maintaining the core features of the adversarial samples, thereby enhancing the defensiveness of the detection model to unseen adversarial samples. An existing diffusion model is improved to be suitable for intrusion detection data, and the data synthesis effect of the diffusion model is improved through adaptive nonlinear noise. Furthermore, an optimal diffusion model is provided, the random noise of the diffusion model adopts self-adaptive nonlinear noise, and the noise level is dynamically adjusted according to time.

The present invention discloses a SAR image ship target recognition method based on the MFF-MA module. Firstly, the expanded and sorted SAR image is sent to the improved VGG model; the SAR image features are fully extracted by improving the first four convolution units of the VGG network, and the output feature map U is obtained through the multi-level feature fusion module; secondly, the feature map U∈R C×H×W The multi-scale attention module is sent to further increase the weight of important features; finally, the output feature maps obtained by the two parallel attention submodules in the MA module are element-by-element added for feature fusion, and then a fully connected layer is used to generate the final classification result. The present invention alleviates the problem of data imbalance, fully mines rich deep-level information through the MFF module, reduces the negative impact of "too high inter-class similarity" of ship targets through the MA module, and models the semantic dependency relationships in the spatial dimension and channel dimension respectively, thereby improving the feature discrimination ability.

The application discloses an Android malware family classification method and system based on an unbalanced heterogeneous graph, and relates to the technical field of computer security and malware detection. An unbalanced isomorphic graph network under each meta path is acquired; model confidence and structure confidence of each node in the unbalanced isomorphic graph network are calculated to select anchor nodes and auxiliary nodes from all nodes in the unbalanced isomorphic graph network; an adaptive mixing coefficient is calculated according to feature vectors of the anchor nodes and the auxiliary nodes, and the feature vectors of the anchor nodes and the auxiliary nodes are weighted according to the adaptive mixing coefficient to obtain feature vectors of new nodes in the unbalanced isomorphic graph network; neighbor nodes of the new nodes are determined through a graph diffusion technology, and the unbalanced isomorphic graph network after adding the new nodes is determined as a balanced isomorphic graph network; and a classification result of a malware family to which each Android application belongs is determined according to the balanced isomorphic graph network of each meta path. The method improves the classification accuracy of Android malware families.

The invention belongs to the technical field of irregular refractory material overturning, and particularly relates to a device and method for achieving irregular material overturning by changing the structure of an overturning machine. Round wheels are arranged at the two ends of the turnover machine, an elliptical wheel is additionally installed on one side of the turnover machine to control the occupied length of the hanging belt, and a groove is machined in the elliptical wheel and connected with the rotating shaft through a key. When an irregular material is turned over, the hanging belt at the small end of the material is placed in the elliptical wheel, and meanwhile, the long shaft of the elliptical wheel is rotated horizontally to supplement the length of the hanging belt, so that the phenomenon that the gravity center of the material shifts due to the change of the length of the hanging belts on the two sides during turning over is avoided. When regular materials are turned over, the round wheels on the two sides can be used, the turnover machine can be used for multiple purposes, and the use efficiency is improved. By changing the structure of an existing turnover machine, the gravity center of an irregular material does not shift in the turnover process, and the technical problem that the gravity center of the irregular material shifts in the turnover process, and consequently the material is damaged is solved.

The present invention provides a strain of Lactobacillus plantarum L-27-2 and an application thereof. The Lactobacillus plantarum L-27-2 of the present invention can effectively alleviate dyslipidemia, intestinal flora imbalance and weight gain caused by a high-fat diet, reduce cholesterol absorption, has a bactericidal effect on common pathogenic bacteria, can be widely used in the addition of probiotics to cat food or the preparation of health products for cats, and the strain has broad market application potential.

The invention discloses a power transmission line inspection image target detection method and system and a computer readable medium, and relates to the technical field of image processing, and the method comprises the following steps: employing a random probability resampling strategy to generate a resampling data set for a power transmission line inspection infrared image data set C in a fine rain weather, and obtaining a resampling data set RC; performing image augmentation on the resampling data set RC by adopting a multi-scale adaptive hybrid strategy to obtain a to-be-trained data set TC; training a target detection algorithm by using the to-be-trained data set TC; and using the trained target detection algorithm to carry out target detection on the power transmission line inspection infrared image in the fine rain weather. According to the invention, accurate detection of the transmission line inspection infrared image target in the fine rain weather can be realized.

This invention relates to the field of image processing, specifically providing an image segmentation processing method, control device, and readable storage medium, aiming to solve the problem of how to achieve more accurate segmentation of subtle categories in images. To this end, the image segmentation processing method of this invention includes: extracting features from the image to be segmented to obtain extracted features and stitched features, wherein the stitched features are obtained by further extracting features from the extracted features; and selecting, according to preset conditions, to perform image segmentation using either the stitched features or the combined stitched features and extracted features, thereby obtaining a segmentation category map. This solution preserves the high resolution of the image and improves the accuracy of image segmentation. During segmentation, in addition to using the stitched features output by the feature extraction network, feature information from the feature extraction stage is also incorporated, providing more referenceable feature information when optimizing the segmentation results, further ensuring the refinement of the segmentation results.

The invention belongs to the technical field of data mining, and particularly relates to a new energy automobile battery multi-fault early warning method based on unbalanced data, which comprises the following steps: acquiring new energy automobile battery monitoring data to be early warned, and inputting the new energy automobile battery monitoring data into a trained new energy automobile battery multi-fault early warning model, a new energy automobile battery multi-fault identification result is obtained; the new energy automobile battery multi-fault early warning model comprises a feature enhancement module, a prediction module, a sample minority mixing enhancement module, a label fine granularity management module and a label correction module. According to the method, data enhancement and loss correction strategies are combined, and the robustness and accuracy of the model to new energy automobile battery fault tag imbalance and noise data are remarkably enhanced.