321 results about "Sample Weight" patented technology

The invention discloses a vehicle surrounding structure fault prediction maintenance method and system based on machine learning, and relates to the technical field of vehicle intelligent monitoring and prediction maintenance. Through fusion of vibration, strain and environment corrosion data and a finite element simulation stress map, an environment-load collaborative damage effect is quantified through a physical degradation model, and a multi-modal feature vector is generated. Constructing a dynamic graph topology network, adjusting edge weights in real time, and synchronously capturing spatio-temporal mechanics characteristics by using spatio-temporal graph convolution; and further identifying key risk nodes through a graph attention mechanism, simulating a damage propagation path in combination with a corrosion attenuation coefficient and a graph diffusion model, and generating a probabilistic fault thermodynamic diagram. The fault prediction precision of the vehicle surrounding structure is improved, an adversarial generative network is constructed to optimize the sample weight, and a quantile maintenance list is output in combination with a physical constraint loss function. Fault prediction and maintenance optimization of the vehicle surrounding structure can be effectively carried out, the service life of the structure is prolonged, and the maintenance cost is reduced.

The invention discloses a slab surface crack prediction method based on PSO-MXGBoost and a prediction model construction method, and belongs to the technical field of surface defect prediction. According to the method, on the basis of a metallurgical mechanism embedded MXGBoost algorithm, a solidification shrinkage rate dynamic correction item is implanted into a traditional XGBoost loss function, the sample weight is adjusted according to a shrinkage rate coefficient calculated according to steel grade components, a theoretical reference value of crack tendency is quantified through a thermodynamic equation, and prediction output of XGBoost is restrained to conform to a continuous casting solidification heat transfer mechanism; meanwhile, key hyper-parameters of the MXGBoost model are optimized through a PSO algorithm, so that an optimal crack prediction model is obtained. According to the method, the prediction precision of the slab surface cracks can be effectively improved, meanwhile, the hyper-parameters of the MXGBoost model are automatically adjusted through the PSO algorithm, manual intervention can be reduced, the training efficiency and prediction accuracy of the model are improved, and therefore multi-target collaborative optimization of model precision and metallurgical feasibility can be achieved.

The invention provides a noise robust text-to-image pedestrian retrieval method and device based on a soft label. The method comprises the following steps: calculating cosine similarity based on image global features and text global features, and normalizing the cosine similarity to generate a soft label representing image-text pairing confidence; distributing a sample weight for each training sample according to the soft label, and obtaining a joint weight for current iteration in combination with a dynamic weight factor progressively increased along with the training process; respectively constructing cross-modal contrast learning loss and similarity distribution matching loss by using the joint weight, and carrying out weighted summation on the cross-modal contrast learning loss and the similarity distribution matching loss to obtain a total loss function; and updating parameters of the image encoder and the text encoder by using the total loss function until training convergence, and obtaining a cross-modal alignment model. According to the method, robust cross-modal alignment can be realized, and the pedestrian retrieval accuracy in a noise scene is improved.

The invention discloses a drainage basin ecological anomaly monitoring method based on clustering processing. The method comprises the steps of data acquisition, drainage basin ecological mapping, radius optimization, initial clustering center selection, drainage basin ecological data clustering processing and drainage basin ecological anomaly monitoring. The invention belongs to the field of ecological monitoring, and particularly relates to a clustering processing-based drainage basin ecological anomaly monitoring method, which comprises the following steps of: introducing a time weight coefficient to construct a time fusion feature vector, and improving the sensitivity to an emergency; the sampling density is automatically adapted by iteratively adjusting the radius, and the monitoring accuracy is improved; introducing a time change factor, constructing a composite index, selecting a core reference point, and objectively emphasizing the most representative core position; the sample weight is introduced to strengthen anomaly monitoring, the attention degree on different indexes is continuously adjusted through core reference point guidance, and the influence of important ecological variables is improved; the weight is updated based on the index stability, the contribution degrees of different indexes to anomaly monitoring are distinguished, and then the anomaly monitoring effect is improved.

The invention discloses a matching threshold adjustment method and system oriented to post portrait dynamic update, particularly relates to the technical field of intelligent recruitment, and is used for solving the problem of systematic offset caused by the fact that low-quality candidate data pollutes a model due to dynamic threshold adjustment of an existing recruitment platform. Candidate skill test scores and performance data corresponding to target post skill labels are obtained; calculating the KL divergence value of the skill test score distribution and the post reference curve; triggering a PID control algorithm based on a KL divergence threshold to generate a threshold correction coefficient; dynamically updating the matching threshold and generating a threshold range; adjusting a sampling weight strategy according to the deviation direction; and the model is retrained to output a recommendation list, collaborative optimization of recruitment efficiency and model stability is realized, and interference of data quality degradation on an algorithm benchmark is effectively inhibited.

The invention relates to the field of intelligent early warning of guaranteed power supply, and provides an abnormal early warning method and system for a guaranteed power supply system, and the method comprises the steps: obtaining the topological structure, real-time operation data and equipment state data of each node in a guaranteed power supply network, employing the conditional entropy and instantaneous causal entropy calculation technology, precisely evaluating the causal intensity between variables, and achieving the early warning of the abnormal state of the guaranteed power supply system. And a causal adjacency matrix dynamically changing along with time is constructed to reflect a complex causal relationship in the network. Then, in combination with an equipment topological structure and dynamic causal information, multi-scale time sequence features are extracted through feature updating and expansion causal convolution, and a comprehensive feature vector is formed; after the vector is input into the Bayesian neural network, a plurality of prediction value samples are generated through sampling weight distribution, a mean value is used as a prediction result, and the prediction uncertainty is measured through variance. Accurate prediction and timely early warning of potential anomalies of the power supply system are realized, and the stability and safety of the system are effectively improved.

The invention relates to the technical field of structure health monitoring, in particular to a steel structure elevator shaft settlement prediction method based on an LSTM (Long Short Term Memory) model, which comprises the following steps: analyzing three-axis dip angle data, screening a dip angle change effective section, extracting three-dimensional displacement data to identify a dip angle linkage mode interval, identifying offset amplitude and constructing a sample weight distribution table; and detecting a jump point, setting a prediction path to obtain settlement prediction information, and comparing residual error identification time migration to generate structure prediction migration trend data. According to the method, the effective data segment is extracted by identifying the dip angle inflection point and combining the three-axis change amplitude difference, the direction accuracy of structural trend judgment is enhanced by combining the main shaft displacement direction identification, the weight of the training sample is adjusted by combining the data offset amplitude, the learning ability of the model for the key deformation time period is optimized, and by utilizing the jump density, the structural trend judgment accuracy is improved. A prediction path starting point is set, behavior responsiveness of trend prediction is improved, and interpretability of a prediction result is enhanced in combination with translation state residual fluctuation analysis.

The invention provides a brain tumor classification method based on deep learning, and the method comprises the steps: uniformly segmenting a multi-mode magnetic resonance imaging image into a plurality of small blocks, generating an enhanced puzzle through random geometric transformation, inputting the enhanced puzzle into a ResNet50 backbone network, extracting primary local features, and optimizing feature distribution; multi-scale feature maps of different stages are obtained, lightweight convolution blocks are constructed, and local pathological relevance is enhanced; performing full-connection layer splicing on the multi-scale feature maps in different stages, introducing a learnable weight matrix to dynamically allocate weights in each stage, and generating fusion features; a progressive parameter unfreezing mechanism is adopted to ensure orderly learning of the model from local to global; the sample weight is dynamically adjusted according to the category frequency, and the local discriminant force and the global consistency are balanced in combination with multi-stage classifier loss weighted fusion; and a final classification result is generated through multi-stage prediction probability weighted average and dynamic threshold adjustment, and clinical availability is improved in combination with a sigmoid calibration module.

The invention discloses a diffusion model optimization method and system based on U-Net parameter enhancement, and relates to the technical field of diffusion model optimization. The method comprises the following steps: acquiring a data set formed by image and text pairs; constructing a diffusion model, and introducing a learnable mask into the diffusion model for sampling the weight from the U-Net structure; current computing resources are obtained, fine tuning strategies are selected according to the computing resources, and the fine tuning strategies comprise the fine tuning strategy based on training and the fine tuning strategy based on a reward model; and pre-training the diffusion model by using the data set, and optimizing the learnable mask based on the selected fine tuning strategy in the pre-training process to complete optimization of the diffusion model. According to the method, the generalization ability of the pre-training model is kept by avoiding updating U-Net parameters.

The invention relates to the technical field of data management analysis, in particular to an indoor radon concentration detection management method and system, and the method comprises the steps: determining that the system is in a normal operation state; obtaining the number of carbon boxes in the current detection batch, the first weight of each carbon box and the second weight after the radon concentration detection sample is put; determining whether the radon concentration detection sample is abnormal according to the first weight and the second weight; if not, enabling the system to sleep for a preset time period T, and detecting the indoor radon concentration; and after the indoor radon concentration detection is finished, obtaining detection data, and generating a data list containing the current batch number, the detection sample ID and the detection data. According to the indoor radon concentration detection management system and method, the system state information, the sample batch information and the sample weight information can be managed in a unified mode by constructing the indoor radon concentration detection management system and method of the self-control detection process, full automation of indoor radon concentration detection is achieved, the indoor radon concentration detection cost is reduced, and the detection efficiency and quality are improved.

The invention discloses a universal domain adaptive image classification method based on distance entropy weighting, and the method specifically comprises the steps: 1, obtaining source domain data and target domain data, 2, building a model, and carrying out the pre-training; step 3, mapping the source domain data and the target domain data to a hypersphere; 4, setting all prototypes of the source domain as 0; step 5, calculating a source domain category weight # imgabs0 # and a target domain sample weight # imgabs1 #; step 6, performing adversarial training by using a weight weighting training model and a domain discriminator D, and determining whether the data is from a source domain or a target domain; obtaining weighted confrontation loss; and 7, screening a target domain sample to obtain a target domain auxiliary domain, and outputting an image for classification. According to the universal domain adaptive image classification method based on distance entropy weighting disclosed by the invention, the problem that image classification is not clear due to the fact that a model in the prior art cannot effectively distinguish and recognize private categories of a target domain is solved.

The invention discloses a convolutional neural network training method for label noise image processing, electronic equipment and a storage medium, and belongs to the technical field of label noise image processing. In order to improve the stability of the training process and improve the label noise filtering performance, the time average neural network is introduced, a more stable sample weight is generated, the stability of the training process is improved, and the label noise filtering performance is improved. For the identified noise data set, label correction is carried out by using the time averaging neural network, and the stability of the label noise correction method is improved because the mutual influence of the deep neural network and the label noise correction part is decoupled. On the whole, the robustness and generalization ability of the model on the data with the noise label are effectively improved. Compared with a traditional method, the method shows remarkable superiority when the label noise problem is processed.

The invention discloses a sleep stage classification method based on multi-scale feature contrast learning. The method comprises the steps that electroencephalogram signals to be classified are acquired; preprocessing the electroencephalogram signal to obtain an enhanced EEG signal; inputting the EEG signal into the trained multi-scale feature comparison representation learning sleep network model, and outputting a final classification result by using the model; the multi-scale feature comparative representation learning sleep network model comprises a comparative representation learning module, a pyramid time context learning module and a classification module which are connected in sequence; a difficult sample mining unit is arranged in the comparison characterization learning module and is used for mining boundary samples and samples with low classification accuracy from a training data set, introducing a sample weighting mechanism into a supervised comparison loss function, setting weights of the samples and limiting the range of the weights through a maximum weight limiting mechanism; and finally, optimizing network parameters in the contrast representation learning module through a supervised contrast loss function.

The invention discloses a multi-agent medical inquiry system based on an improved GRPO algorithm, and belongs to the field of intelligent medical treatment. According to the method, firstly, a new medical data set containing illusion answers is constructed through an open source medical data set, clinical correlation of the data set is evaluated through multiple large models, and weights are given to samples; then, an improved GRPO algorithm is adopted to optimize the multi-modal large model, and the model is encouraged to explore different answers by progressively decreasing KL divergence constraints in the optimization process; the optimized model is applied to construction of a medical question and answer system, the system works cooperatively through multiple agents including a user agent, a hospital guide agent, a doctor agent and an expert agent, and automatic illness state analysis, consultation department judgment, medical opinion generation and specialized correction of an inquiry result are achieved. According to the invention, the accuracy and specialty of medical inquiry are effectively improved, the misdiagnosis rate is reduced, the medical efficiency is improved, and more accurate and comprehensive inquiry services are provided for users.

The invention relates to an embankment slope stability assessment method based on artificial intelligence, and belongs to the technical field of embankment slope monitoring and assessment. The method comprises the following steps: acquiring and marking embankment slope stress sensing data; dividing the data into a plurality of spatio-temporal data blocks; constructing a stability evaluation model; extracting multi-scale convolution features by adopting multi-scale cavity convolution of a high-frequency channel and pooling-deconvolution operation of a low-frequency channel to obtain a fused multi-scale feature matrix; a hidden state sequence is obtained through a space attention mechanism and a double-door-setting mechanism; calculating time interval saliency based on the hidden state vector, then calculating a weighted feature vector, further obtaining a weighted feature matrix, and processing through deep convolution and point-by-point convolution to obtain a pooling feature vector; carrying out stability evaluation grade classification through learnable category prototype and gating feature transformation; and dynamically adjusting sample weight and constraint attention distribution by adopting a total loss function. According to the method, the progressive instability identification capability can be improved.

The invention provides a data security encryption method and system based on high-security machine learning, and the method comprises the steps: collecting an access behavior of to-be-encrypted data, and calling context and environment meta-information to construct a structured feature sequence; establishing index mapping between the feature sequence and a historical encryption execution record, and extracting significant features by using a feature scoring and information gain evaluation method; the system executes multi-level decision path search based on hierarchical condition division according to the significant feature sequence of the current task, and judges the security level and the encryption risk situation of data; the system constructs an encryption control sequence including a parameter instruction block and an execution logic structure according to the task risk level and the strategy index; the encryption engine analyzes the instruction packet and executes a corresponding encryption task, and generates a complete record file containing metadata and a tracking tag at the same time; and the system performs statistical analysis on decryption feedback logs and encryption abnormal records, and optimizes the strategy dictionary by adopting a sample weight updating and feature re-evaluation method.

The invention discloses a voice data processing method and device based on a large model, a storage medium and an electronic device, and relates to the technical field of voice data processing, and the method comprises the steps: carrying out the data amplification operation of labeled data of a labeled data set, and obtaining an amplified training data set; classifying the amplification training data set into a positive sample data set and a negative sample data set by using the analytical model, and generating a preference optimization data set by using the positive sample data set and the negative sample data set; generating a weighted loss function according to the sample weight of the preference optimization sample and the original loss function of the analytic model, obtaining a data amplification model for completing a preset training target by using the weighted loss function, and performing data amplification on newly input text data by using the data amplification model to obtain a target amplification data set. The technical problem that the quality of data obtained by an existing data amplification method is low is solved.

The invention discloses a hyperspectral image classification method based on frequency domain denoising and element gradient correction, and the method comprises the following steps: carrying out the preprocessing of all hyperspectral image data, and dividing an overall training sample set formed by the processed hyperspectral images into a training set and a verification set; constructing a sample weighting model based on frequency domain denoising and element gradient correction; in the training process, a parameterization frequency spectrum gating sensing transformation module is utilized to map features to a frequency domain through discrete Fourier transform, a learnable frequency spectrum response function is utilized to adaptively suppress spectrum jitter noise, and finally pure features are reconstructed. And automatically constructing a high-confidence pseudo-clean verification set based on a Gaussian mixture model and time domain consistency. According to the method, a time domain momentum updating mechanism is introduced, the variance of statistical estimation is effectively smoothed, random interference caused by training fluctuation is resisted, and the accuracy of pseudo clean set construction and the convergence stability of overall model training are further improved.

The invention discloses an efficient graph comparison learning method for social recommendation. As an emerging self-supervised learning normal form, graph contrast learning is excellent in response to data sparseness and cold start due to the fact that the graph contrast learning can effectively capture similarity and heterogeneity characteristics in a graph structure, although the learning normal form achieves a good effect in a recommendation system, the graph contrast learning can be used for solving the problems of data sparseness and cold start. However, the method still faces three defects: (1) average neighbor aggregation and a non-adaptive representation reading mechanism are adopted in a message propagation process, and high-quality node representation is difficult to learn; (2) a visual angle is enhanced by depending on a random disturbance generation graph during intervention of comparative learning, which may destroy the inherent structure of graph data and further weaken the accuracy of the model; and (3) equally treating all observation samples during parameter optimization, and neglecting the difference influence of positive samples in different training stages. Specifically, aiming at the problems, the invention provides an efficient graph contrast learning method (EGCL for short). The method comprises the following steps: firstly, designing a graph adaptive propagation module, improving an information propagation rule of a graph neural network by referring to a thermonuclear thought and an attention mechanism, and realizing differentiated aggregation of neighbor nodes by adopting a learnable weight distribution strategy; secondly, designing a double contrast learning normal form which does not need graph enhancement, and realizing mutual promotion of node characterization through intra-domain contrast learning (inter-CL) and inter-domain contrast learning (inter-CL); and finally, introducing a sample weight adaptive efficient optimization algorithm, converting the training process into a double-layer optimization problem, and adaptively adjusting the contribution degree of each sample to model optimization in different stages.

The invention relates to the technical field of data processing, in particular to a transaction security authentication method based on multi-source information fusion, and the method comprises the steps: collecting a current month training fraud sample and a current month normal transaction sample, obtaining the evolution degree of each training fraud sample in the current month according to the difference between the feature vectors of the training fraud samples in the current month and the feature vectors of the historical fraud samples in the historical time periods before the training fraud samples in the current month, and obtaining the growth degree of each training fraud sample in the current month; obtaining the sample weight of each training fraud sample in the current month according to the growth degree and the evolution degree, endowing the training fraud sample in the current month with the corresponding sample weight, inputting the training fraud sample in the current month and a normal transaction sample in the current month into a LightGBM model for training to obtain a trained LightGBM model, and identifying the fraud behavior of the latest transaction record according to the trained LightGBM model. According to the invention, the accuracy of fraudulent behavior identification is improved.

The invention discloses an uncertainty guided few-sample harmful speech detection method (U-GIFT). According to the method, a pre-training language model is finely adjusted based on a small number of labeled samples, and a semi-supervised self-training and uncertainty guiding strategy is combined. Monte Carlo Dropout is started in the reasoning stage, multiple times of random forward propagation are carried out to obtain sample posterior distribution, prediction entropy and information gain are calculated, pseudo-label samples are sorted and screened, and only high-confidence samples are selected to be added into a training set. And in order to reduce the influence of a pseudo labeling error, designing a stability weighting mechanism, giving a sample weight according to a prediction variance, and constructing a joint loss function, so that the model preferentially learns a stable sample to improve the detection performance. According to the method, the semantic and attention mechanism of the pre-training model is utilized, the detection effect is remarkably improved under the conditions of few samples, imbalance, multiple languages and cross domains, models such as BERT, RoBERTa, XLM-R, LLaMA2 and DeepSeek-R1 are compatible, and the method is suitable for content auditing and risk prevention and control.

The invention relates to the technical field of storage garbage recycling algorithms, in particular to a garbage recycling system based on a genetic algorithm, which comprises an environment sensing module, a dynamic excitation pool module, a gene coding module, a genetic operation module, a strategy evaluation module, a cold and hot data verification bridge module, a wear leveling compensator and an iterative deployment module. The environment sensing module is used for scene feature extraction and load mode classification. Through dynamic parameter optimization and a multi-scene adaptation mechanism, an optimal strategy parameter combination is automatically searched by utilizing a genetic algorithm, and controllable disturbance is introduced by dynamically adjusting sample weight and polynomial variation in combination with an entropy weight method, so that multi-target balance of write amplification, block efficiency and wear balance is realized; meanwhile, through a cold and hot data verification bridge and a wear leveling compensator module, the data state transition trend is predicted, parameter deviation is actively corrected, high robustness and long-term stability can still be kept in a complex load scene, and finally a closed-loop self-adaptive garbage collection strategy generation system is formed.

The invention discloses a crop disease and insect pest detection algorithm fusing self-attention and a sample weighting mechanism. A YOLOv8 network is used as a basic model; through a two-branch self-attention DF-MSA architecture, the ability of the model to acquire pest target position information is enhanced; a cross-layer feature fusion module is introduced into the feature pyramid network structure, and multi-scale information is fused; and a sample weighting function is adopted to reduce the influence of difficult sample imbalance on a detection result, so that the detection accuracy is improved. And inputting the trained IP-YOLO by using a test set sample, and outputting a result. The method has high stability and robustness, and an effective means is provided for improving the performance of a crop disease and pest detection system.

The invention relates to a multi-dimensional degradation control method and system based on self-encoding and fuzzy modeling, and is suitable for health management and life prediction of industrial equipment. The system collects equipment degradation data through multiple sensors, constructs a three-dimensional tensor and performs data fusion and weight optimization by using CP decomposition and an entropy weight method; detecting a multi-dimensional collaborative change point by adopting a mahalanobis distance and a combined likelihood ratio; low-dimensional features are extracted through an auto-encoder, and degradation modes are divided in combination with a clustering algorithm; establishing a joint degradation equation fusing a drift term matrix, diffusion covariance and fractal Brownian motion, and depicting long-range dependence; on the basis of fuzzy Markov chain modeling state transition fuzziness, a multi-dimensional degradation path is generated in combination with a Copula-Monte Carlo algorithm, the sample weight is dynamically adjusted, and the residual life distribution is estimated and predicted by using kernel density. According to the method, the degradation modeling precision and working condition adaptability of the complex industrial equipment are improved, and technical support is provided for precise life prediction and health management.

The invention is applied to the technical field of bearing fault diagnosis, and particularly discloses a multi-source-domain unsupervised-domain adaptive bearing fault diagnosis method, which comprises the following steps: S1, collecting vibration signals under different working conditions, and carrying out a preprocessing step to respectively obtain a plurality of source domain data sets and target domain data sets; and S2, establishing a network of the bearing fault diagnosis model. According to the multi-source-domain unsupervised domain adaptive bearing fault diagnosis method, a larger weight is given to a source domain sample with high similarity with a target domain through a sample weighting mechanism, and meanwhile, the confusion degree of the target domain sample is quantized and the weight is adjusted by means of prediction entropy; the method can effectively reduce the interference of dissimilar samples and easily-confused samples on the bearing diagnosis process, calculates the difference between a source domain and a target domain through the Worsteesson distance in combination with a multi-source domain information fusion module, and distributes the weight, thereby achieving the weighted fusion of multi-source domain classification results.

The invention discloses a federated backdoor defense method based on decoupling contrast learning, and the method comprises the steps: training a backdoor model based on a backdoor sample, and immediately stopping training after the backdoor model converges on the backdoor sample; respectively extracting a penultimate layer vector of the backdoor model and the local model from a sample pair held by the malicious client as a backdoor feature and a clean feature; comparing and learning the separated back door features and the clean features, and learning the clean features for the local model by using a sample weighting strategy to train the local model to obtain a trained local model; and sending local model parameters of the trained local model to a global server, and generating model parameters of a new global model based on the local model parameters through an aggregation function. The method aims at reducing information dependence between backdoor features and clean features through comparative learning, so that local model learning is free of backdoor representation, and the robustness of a global model is improved.

The invention discloses an insurance fraud risk real-time evaluation method based on a dynamic Bayesian network. The method comprises the following steps: S1, collecting time sequence behavior data of an insurance user; s2, performing preprocessing to generate a structured observation sequence; s3, dividing into continuous time slices according to time, and constructing a time sequence sample; s4, constructing a dynamic Bayesian network; s5, calculating a posterior probability of a state variable in each time slice, and generating a sample weight; s6, executing an improved increment EM algorithm based on the sample weight, and optimizing dynamic Bayesian network parameters; s7, monitoring posterior probability entropy change of the state variable, and performing structure adjustment; and S8, the reasoning process is executed again, and a risk scoring report is generated. According to the method, the dynamic Bayesian network and the incremental EM algorithm are fused, the insurance fraud risk time sequence evaluation model is constructed, and the method has the advantages of being high in real-time performance, self-adaptive in structure, interpretable in result and the like.

The invention provides a low-dielectric-constant material dielectric property prediction method and system based on machine learning, and the method comprises the steps: obtaining material data, and carrying out the preprocessing to obtain basic attributes; performing feature engineering to obtain a feature matrix; dividing the feature matrix into a data set and processing the data set; training the model to obtain an optimal model; and evaluating the optimal model, and serializing and storing the optimal model. The structure of a machine learning method combining specific feature engineering, training data noise enhancement, a sample weighting strategy and a random forest model solves the problem that an existing machine learning model accurately predicts a specific low dielectric constant (lt; and 5) the problems that the precision in the aspect of materials is insufficient and the efficiency of a traditional material screening method is low are solved, and the efficiency and accuracy of discovery of the low-dielectric-constant new materials and the robustness and generalization ability of prediction are 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 provides a water supply network water leakage abnormity identification method based on NMF dimension reduction, and relates to the technical field of water supply network abnormity identification, and the method comprises the steps: embedding hydrophone sample data containing expert labeling information into a sample weight matrix as a row vector, and constructing a sample dimension loss item; hydrophone sample data containing industry feature priori are mapped to the feature basis matrix to serve as rows, and feature dimension loss items are constructed; combining the two loss items to construct a total loss function and solving the total loss function to obtain an optimal sample weight matrix; and performing clustering analysis based on the optimal sample weight matrix to obtain a clustering affiliation result of the hydrophone sample data so as to judge whether the water supply network to be identified has water leakage abnormity or not. According to the method, the double-constraint NMF framework fusing the expert knowledge of the sample dimension and the industry priori knowledge of the feature dimension is constructed, the accuracy and interpretability of hydrophone feature dimension reduction are considered, and the problems that an existing method is poor in interpretability, less in priori knowledge utilization and the like are solved.