370 results about "Decision boundary" patented technology

Disclosed in the present invention are an adaptive deep transfer fault diagnosis method and system, an apparatus and a medium. The method comprises the following steps: S1: collecting vibration acceleration signals of industrial equipment under different working conditions, and dividing same into a source domain data set and a target domain data set; S2: building a self-tuning universal domain adaptive fault diagnosis model, which comprises a shared feature extractor, a known classifier and a plurality of unknown classifiers; S3: separately calculating a classification loss of known faults of the source domain, a discriminative loss of the plurality of unknown classifiers, a target domain soft consistency regularization loss and an information entropy maximization loss; S4: introducing a dynamic weighting strategy based on model uncertainty assessment to optimize the model parameters; and S5: using the model for diagnosis. The present invention can fully mine valid information in data, can establish reliable class decision boundaries, and in addition, uses the self-tuning dynamic update strategy to adjust weightings corresponding to different loss functions, thus allowing for quick generalization of the model to different industrial diagnosis scenarios.

The invention provides a hyperspectral image open set spectral spatial feature extraction and classification method, and aims to improve the classification precision of a known category and an unknown category in a hyperspectral image and the robustness of a classification model. According to the method, input data are preprocessed through fractional Fourier transform, and rotation analysis of a signal time-frequency plane is achieved. Multi-scale features of multi-branch cavity convolution are fused through an enhanced spectrum space residual module, deep separable convolution enhancement nonlinear representation of a lightweight convolution enhancement block is combined, and self-adaptive separation and weighted fusion of high / low frequency features are realized by innovating a dual-frequency enhancement module. Proposing a class perception comparison loss function, integrating anchoring loss, triple loss and regularization terms, and collaboratively optimizing intra-class compactness, inter-class separability and an open set decision boundary. The model framework provided by the invention is combined with the improved loss function, so that the classification precision of the known class and the unknown class of the hyperspectral image is remarkably improved.

The invention discloses a hydropower station workshop deformation monitoring method, which comprises the following steps of collecting multi-dimensional monitoring data and point cloud data of a hydropower station workshop, performing adaptive filtering and noise reduction on the multi-dimensional monitoring data, performing spatial registration on the multi-dimensional monitoring data and the point cloud data in a unified three-dimensional coordinate system, constructing a multi-channel time sequence deep learning model, and performing multi-channel time sequence deep learning on the multi-channel time sequence deep learning model; a dynamic early warning threshold model is constructed based on material characteristics, equipment operation parameters and historical monitoring data, risk assessment is carried out, a three-level response mechanism is triggered according to a risk assessment result, a sensor reliability evaluation index is established based on deviation analysis of monitoring data and an early warning result, and a data fusion weight is dynamically optimized. Adjusting the decision boundary of the classification model, and carrying out visualization and traceability analysis on the multi-dimensional monitoring data; the invention further discloses a hydropower station plant deformation monitoring system. According to the invention, a multi-dimensional and three-dimensional risk assessment system is established through data monitoring and real-time analysis, and the structural safety of the hydropower house is guaranteed to the maximum extent.

The invention discloses a keyboard key defect detection method and system based on machine vision, and the method comprises the following steps: collecting the multi-modal data of keyboard keys, and carrying out the fusion preprocessing of reflection component separation, dynamic gamma correction and multi-modal constraint alignment; constructing a double-flow deep neural network for defect detection, wherein the first branch network adopts an improved U-Net architecture embedded with a CBAM attention module to extract texture features; the second branch network adopts a PointNet + + architecture to process three-dimensional geometrical characteristics; realizing cross-modal feature association through a feature fusion layer, wherein a fusion weight is adaptively adjusted according to a material type; and outputting a detection result based on the cascade classifier, wherein the method comprises the following steps: positioning a suspected defect area by a first-stage YOLOv5 network; the second-stage ResNet50 network is used for completing defect classification; and the dynamic threshold segmentation algorithm adjusts the judgment boundary according to the material type. According to the invention, high-precision and high-efficiency detection of keyboard key defects is realized.

A natural gas pipeline multi-working-condition fault diagnosis method and system based on Bayesian adversarial attack and single-source domain transfer, relating to the technical field of mechanical fault detection and diagnosis. The core of the method is using the transfer learning technology to solve the problem of insufficient generalization ability of existing deep reasoning models when processing pipeline fault diagnosis tasks under different working conditions. The method mainly comprises the following steps: constructing an attack sample generator on the basis of a Bayesian network, wherein the attack sample generator is used for generating, by adding delicately designed tiny disturbance into an input sample, an attack sample that can cause an reasoning model to make an incorrect decision, so as to mine and analyze a defect of the reasoning model; constructing a domain discriminator on the basis of the Bayesian network, wherein the domain discriminator is used for assist in generating a high-concealment attack sample by means of adversarial learning between the domain discriminator and the generator, that is, there is almost no visible difference between the high-concealment attack sample and an original sample; and constructing a classifier on the basis of the Bayesian network, and by expanding the distance between the attack sample and an original decision boundary of the reasoning model, constraining the posterior distribution of network parameters of the reasoning model to be adjusted towards a higher score of the attack sample, thereby enhancing the adaptability and robustness of the model when facing disturbance in different domains. By means of the steps, the present invention effectively solves the problem of missing reporting and false reporting risk improvement caused by poor generalization ability of traditional deep learning models under different working conditions.

The invention discloses a method and a system for automatically classifying heterogeneous data based on business deep learning, particularly relates to the technical field of data classification, and is used for solving the problems of limited classification accuracy and insufficient interpretability caused by cross-modal feature entanglement in the existing method. The method comprises the following steps: decoupling and separating an independent feature and a redundant feature vector through a modal implicit feature, generating a noise correction instruction based on orthogonality residual evaluation, carrying out cross-modal noise distribution consistency correction on the redundant feature, quantifying a gradient direction conflict rate in combination with a vector space projection relation, and dynamically adjusting a back propagation path. Fusion path switching is triggered through semantic matching degree and contribution degree balance monitoring, a self-adaptive classification decision boundary function is constructed, and high-precision classification of heterogeneous data and strong generalization adaptation of service scenes are achieved.

The invention provides an image noise mark feature selection method and system, a storage medium and a computer. The method comprises the steps of obtaining a to-be-processed image noise mark data set; embedding a sample set in the image noise mark data set into a multi-granularity fuzzy cluster to construct a dynamic fuzzy membership evaluation matrix; dynamically evolving a multi-level high-precision granular ball cluster; obtaining mark distribution with high identification degree; constructing a rough perception feature evaluation framework based on granular ball topology driving, extracting decision equivalence classes by combining rough set upper and lower approximation and extended positive domain models, and determining and measuring the contribution degree of each feature to a decision system by fusing multi-granular-ball decision boundary information based on a dependency degree quantitative model; a particle and ball structure consistency verification mechanism is introduced, and multi-level evaluation is carried out on the importance of the features through dependency and consistency. According to the method, the optimal feature subset with strong anti-noise performance and high discrimination capability is obtained, and stable and efficient input support is provided for a subsequent image noise mark learning model.

The invention discloses a noise text open intention classification method and system based on a large model, and belongs to the technical field of natural language processing and artificial intelligence, and the method mainly comprises the steps: extracting text features through a pre-trained large language model, carrying out the clustering of the text features through unsupervised granular ball clustering, and constructing intra-class structure representation in a feature space; dividing the training sample into a clean sample, an internal distribution noise sample, an external distribution noise sample and an uncertain sample based on a global particle-ball-level OOD tendency index and a sample-level consistency index; constructing a joint loss function guided by beneficial noise, and performing differentiation training on the model by using a sample division result; and constructing a multi-granularity decision boundary based on a granular ball structure, wherein the multi-granularity decision boundary is used for judging the category of the input sample. According to the method, multi-granularity structure information, a noise pattern recognition mechanism and a joint representation learning strategy are fused, and the recognition robustness and generalization ability of the model in a complex open environment are effectively improved.

The invention relates to the technical field of data processing, in particular to an abnormal mode data processing system driven by power marketing big data, which comprises a distributed collaborative acquisition module for constructing a space-time alignment three-dimensional data stream, a multi-modal feature reconstruction module for separating periodic noise and quantizing environmental interference, and a data processing module for processing abnormal mode data. The resistance feature decoupling module generates a purification feature vector set and a noise confidence index through orthogonal projection, and the dynamic algorithm adaptation module dynamically schedules an isolated forest algorithm, a weighted distance measurement algorithm and a sparse self-encoding clustering algorithm according to the noise confidence index. The behavior chain verification module establishes a combined physical rule verification mechanism of an environment temperature threshold value, a load deviation degree and an equipment state, and the closed-loop strategy engine module adaptively adjusts a feature decoupling loss function weight according to a decision boundary offset, so that the accuracy and the environmental adaptability of real electricity consumption abnormity identification in a complex noise environment are effectively improved.

The invention relates to the technical field of remote sensing image processing, and discloses a remote sensing image adaptive identification method and system for territorial space planning, and the method comprises the steps: obtaining a multi-source remote sensing image data set of a research region, feature extraction, cloud detection, quality evaluation and adaptive preprocessing are carried out; carrying out prototype network coding, calculating a category prototype and probability, and supporting fine tuning of a set; carrying out multi-scale cavity convolution and category scale attention fusion; evaluating the adaptability score of the comprehensive fusion feature map set, and carrying out weighted fusion, classification and normalization; change detection is carried out, stable and change regions are segmented, and time sequence context features are extracted and constrained optimization is carried out; entropy is fused, a boundary is decided, uncertainty is estimated, and weighted fusion is carried out according to a change area; conditional random field optimization, confidence level grading and connected domain identification are carried out; the automation level, the adaptive capacity and the recognition reliability of remote sensing monitoring of territorial space planning are improved.

The invention discloses a large language model enhanced reinforcement learning training method for key automatic driving scene stable decision, and the method comprises the steps: constructing an LLM-based vehicle decision intelligent agent, and providing a high-level guidance strategy through a chain thinking technology and a knowledge experience pool; then, jointly constructing a dynamic intervention mechanism based on the environmental risk sign and the decision boundary parameter, thereby determining an LLM guidance opportunity; finally, through expert guidance algorithm design, the prior knowledge of the LLM is effectively integrated into the deep reinforcement learning strategy network by adopting an adaptive exploration-imitation fusion loss function. The DRL model is better in performance in a vehicle driving decision-making task and shows higher generalization ability in various scenes, and the robustness, the sample efficiency and the overall generalization level of the DRL model can be remarkably improved.

The invention relates to the technical field of information security, in particular to a security baseline setting method and system equipment based on clustering decision and a medium, and the method comprises the steps: collecting original data, carrying out the construction of a feature vector through feature conversion, generating a sample, and marking and extracting the sample; constructing a classification model by using manifold learning, inputting the extracted feature vectors into the constructed classification model, optimizing the model through training, and outputting classification labels; processing the decision boundary by using constraint optimization and feature scaling to obtain a standardized feature vector converted by a decision boundary model, grouping based on the standardized feature vector, and extracting a baseline; the security policy is generated through data analysis, and the data is dynamically updated and continuously monitored, so that the unified configuration and centralized management of the security policy are realized, and the policy implementation efficiency and the system expansibility are remarkably improved.

The invention provides a medical image classification method based on semi-supervised dynamic fusion matching, and belongs to the technical field of images, and the method comprises the steps: generating a pseudo label based on a medical image through calculating the correlation of samples under a weak enhancement view and a strong enhancement view, and carrying out the calibration and fusion processing, and generating a fused class pseudo label; dynamically calculating a threshold value at each time step, and based on dynamic adjustment of the threshold value, optimizing selection and category learning of fused category pseudo labels through a consistency loss function; for long-tail distributed medical image data, category balance loss is introduced, and medical images are classified by dynamically adjusting decision boundaries of each category and taking a total loss function as a target. According to the method, the problems of data scarcity, category imbalance and false label optimization in medical image classification are solved.

The invention discloses a multi-modal open intention recognition method and system based on pellet characterization, and belongs to the technical field of artificial intelligence and multi-modal intention understanding, and the method comprises the steps: carrying out the feature extraction and modal fusion of multi-modal input data; carrying out structural modeling on the feature representation of each mode and the fusion mode through an adaptive particle and ball clustering method, and generating a multi-granularity particle and ball set; the mass centers of the pellets serve as multi-granularity anchor points, and the pellets with the same labels in different modalities are aligned; introducing a weighting mechanism based on purity and sample scale into the fusion mode; generating a boundary-constrained pseudo-distribution outer sample in the fusion modal space; and constructing a self-adaptive decision boundary based on the fusion modal particle ball obtained by training, and carrying out known class classification and unknown class detection. According to the invention, by introducing the multi-granularity anchor point and the structure perception particle-ball representation mode, the joint recognition of the known category and the unknown category in the multi-modal scene is realized, and the accuracy and robustness of intention recognition are remarkably improved.

The invention relates to an abnormity alarm method and system for a feed production control system, and the method comprises the following steps: collecting multi-field data related in a feed production process in real time, the multi-field data at least comprising production equipment operation parameters, material characteristic parameters, environmental parameters and process control parameters; preprocessing the collected multi-field data, and fusing to generate a multi-field description feature vector representing the current production state; inputting the multi-field description feature vector into a pre-trained fault prediction model, wherein the fault prediction model respectively calculates the confidence of the fault type prediction result and the confidence of the fault point prediction result; wherein the confidence coefficient is in positive correlation with the distance from the feature vector to a known fault mode clustering center or the distance from the feature vector to a decision boundary; and when the fault prediction result indicates that an abnormal risk exists and the corresponding confidence exceeds a preset threshold, triggering a corresponding alarm signal.

The invention discloses a privacy protection constrained passive unsupervised graph domain adaptive mechanical fault diagnosis method and device, and the method comprises the steps: inputting a mechanical vibration signal into a pre-trained mechanical fault diagnosis model, and outputting a diagnosis result; the mechanical fault diagnosis model is obtained by training a passive domain multi-scale image domain adaptive network by adopting training data, and the passive domain multi-scale image domain adaptive network comprises a multi-scale feature extractor module, a topological graph convolution module and a feature classifier module which are connected in sequence; the training process comprises two stages of performing pre-training by using source domain data and performing unsupervised domain adaptive training by using target domain data, and a loss function of the training process comprises a label smooth cross entropy loss function for the pre-training process. And the unsupervised domain adaptive loss function used for the unsupervised domain adaptive training process comprises neighbor clustering loss, regularization loss and label prediction diversity loss. The objective of the invention is to realize complete representation and generalization of fault features by deeply mining topological structure information, and optimize decision boundaries to significantly improve inter-domain alignment precision in a passive domain adaptive process, thereby solving the problem of cross-working condition diagnosis of mechanical equipment under data privacy protection constraints.

A system to protect an asset includes a plurality of monitoring nodes each generating a data stream of current monitoring node values in time-domain. An anomalous space data source stores sets of anomalous feature vectors for each of the plurality of monitoring nodes generated by an anomaly detection model. A current system function processor, coupled to the plurality of monitoring nodes over a network, receives receive data streams from each of the plurality of monitoring nodes, and generates a set of current feature vectors. An anomaly detection computer coupled to the plurality of monitoring nodes receives the data streams from the plurality of monitoring nodes and the set of anomalous feature vectors, and outputs at least one decision boundary based on processing, using the anomaly detection model, of the current feature vectors relative to the sets of anomalous feature vectors.

The invention discloses a decision boundary adaptive open set fault diagnosis method based on a domain adversarial neural network, and the method comprises the steps: S1, obtaining a rolling bearing fault data set under multiple working conditions and multiple fault states, and dividing the rolling bearing fault data set into source domain data and target domain data; s2, establishing a bearing fault diagnosis model, wherein the bearing fault diagnosis model can determine a decision boundary of a known fault state; performing iterative training on the bearing fault diagnosis model, and when the total loss function is converged, obtaining a trained bearing fault diagnosis model; the total loss function comprises a Bhatcharyya coefficient loss function, and the Bhatcharyya coefficient loss function is utilized to reduce the feature overlapping degree of different types of faults; and S3, based on the trained bearing fault diagnosis model, carrying out fault diagnosis on unknown faults in the actually collected bearing operation signals. According to the method, cross-working-condition open-set fault diagnosis can be realized, and the fault diagnosis capability is remarkably improved by ensuring accurate identification of unknown faults.

The invention relates to an abnormal transaction behavior recognition method and system based on artificial intelligence, and belongs to the technical field of financial transaction risk control, and the recognition method comprises the steps: collecting real-time transaction flow data, user behavior time sequence data and association graph data of a user, carrying out the space-time alignment processing, and generating a space-time aligned structured data set; dynamically calculating a transaction time attenuation factor; extracting a composite feature vector, inputting the composite feature vector into a pre-trained dual-channel decision model, and performing confidence weighted fusion on a dual-channel decision result based on a time decay factor to obtain a final risk score; dynamically adjusting a risk judgment threshold, and generating a dynamic decision boundary; judging whether the final risk score exceeds a dynamic decision boundary, if so, outputting an abnormal transaction behavior recognition result, and judging a transaction risk level; and triggering a risk disposal action corresponding to the transaction risk level. The method can improve the understanding depth of complex transaction behaviors, and reduces the false alarm rate of abnormal transaction recognition.

The invention relates to the technical field of image processing and traffic safety monitoring, in particular to a dynamic traffic marking inverse reflectivity intelligent monitoring system and a method thereof.According to the system, marking images are collected through a multispectral imaging technology, a precise space mapping relation is established through laser ranging, marking feature parameters are extracted from multi-band images, and the dynamic traffic marking inverse reflectivity intelligent monitoring system is obtained. A multi-dimensional feature matrix including reflection features, space geometry, time change and environmental influence is constructed, a nonlinear dynamic model of a traffic marking state is established based on the multi-dimensional feature matrix, a critical point and a bifurcation point in a degradation process are identified, and the system can calculate the change trend of the inverse reflectivity of the marking and generate multiple possible degradation paths. According to the method, the real-time dynamic monitoring of the inverse reflectivity of the traffic marking is realized, the measurement precision is improved, the prediction accuracy is enhanced, scientific decision support is provided for road maintenance, and the road traffic safety performance is remarkably improved.

The invention discloses a remote sensing knowledge distillation method and system based on dual-mode feature spectrum decoupling, and the method comprises the steps: inputting a remote sensing image, carrying out the parallel extraction of a full-frequency feature map through a teacher model and a student model, and decomposing the full-frequency feature map into a low-frequency component and a high-frequency component through two-dimensional discrete wavelet transform; performing adaptive spatial feature enhancement on the low-frequency component and the high-frequency component through a density-independent spatial weighting mechanism to generate explicit distillation loss; inputting each component feature map output by the teacher model into a corresponding pre-trained knowledge amplifier to generate a thermodynamic diagram, predicting and capturing a decision boundary confidence implied by the teacher model through the thermodynamic diagram, and constructing implicit knowledge constraints based on student feature responses; target detection loss, explicit distillation loss and implicit knowledge constraint are combined, student model parameters are optimized through gradient propagation, and meanwhile, teacher model and knowledge amplifier parameters are frozen to guarantee knowledge migration stability.

The invention relates to the technical field of data processing, in particular to a global unattended intelligent monitoring method based on multi-source data fusion, and the method comprises the steps: obtaining a real-time feature vector of a to-be-recognized target to obtain a global discrimination score; evaluating the category uncertainty of the target to be identified; according to the category uncertainty, a dynamic weight used for model fusion is generated in a self-adaptive mode; and obtaining a local discriminant by using a pre-trained local detail model, carrying out weighted fusion on the local discriminant and the global discriminant by using a dynamic weight to obtain a final discriminant, and determining the category of the to-be-recognized target based on the final discriminant, thereby realizing adaptive recognition of the unmanned aerial vehicle and the birds. Through the technical scheme of the invention, a nonlinear decision boundary of the unmanned aerial vehicle and birds can be effectively constructed, and the recognition accuracy of a fuzzy target in a complex scene is remarkably improved.

The invention relates to the technical field of power system monitoring, in particular to a power distribution terminal fault monitoring system and method. The dual-channel acquisition module is used for acquiring two analog quantities of a power distribution room and mapping the two analog quantities into a power grid environment characteristic matrix; the cyclic mapping module is used for carrying out decoupling operation and generating a noise separation feature vector; constructing a background noise dynamic reference based on the noise separation feature vector, and generating a judgment boundary threshold; performing cyclic increasing and resetting in the fault recall storage unit through a write-in pointer to form a terminal state data stream; the fault judgment module is used for comparing the amplitude of the noise separation feature vector with a judgment boundary threshold value and generating a fault locking instruction; and based on the fault locking instruction, generating a termination signal, freezing the write pointer, and extracting the partial discharge data file from the terminal state data stream. According to the method, a closed-loop monitoring system in which the environment self-adaptive reference and signal feature decoupling support each other is constructed, and a fault backtracking sample with time-space alignment and complete logic is provided.

The invention relates to a risk identification method and system. The method comprises the following steps: acquiring communication behaviors, equipment fingerprints and service interaction data in real time through a privacy compliance interface; carrying out parallel preprocessing and safe desensitization on the data; inputting a feature construction engine to extract a communication mode, an equipment behavior and an interactive semantic feature vector in parallel, constructing a dynamic weighted hypergraph communication map, and outputting a social risk feature vector by using a time sequence hypergraph neural network; inputting the four types of features into a privacy perception multi-mode gating attention fusion module to output fusion features; generating a risk score through a deep neural network classifier; and monitoring an abnormal event based on Apache Flink, adaptively adjusting a decision boundary in combination with the dynamic risk entropy, and triggering secondary verification. According to the method, the problems of multi-modal data conflict, privacy disclosure and decision stiffness are solved, and the recognition precision and the real-time performance are improved.

The invention discloses a new energy power station inspection method based on a multi-modal large model small sample open set, and the method constructs a multivariate text learnable prompt, and guides a detection model to form a finer-grained category decision boundary through the text information related to an aggregation task. Due to the lack of real unknown class samples in the training process, the mining of unknown class pseudo samples is regarded as a bipartite graph matching task for the first time, and the model is optimized to form a compact unknown class decision boundary by adding unknown class virtual nodes, constructing a cost matrix and mining the unknown class pseudo samples. In order to solve the problem that known classes and unknown classes of small samples are prone to confusion, the method proposes unknown class optimization loss based on cost perception, considers the classification and positioning quality of the unknown classes, and improves the open set detection performance of the model. According to the method, transformation from a single-mode vision small model to a multi-mode vision large model is realized, and good generalized small sample open set target detection performance can be obtained only by a small amount of training data.

The invention provides a microphone anomaly detection method based on multi-mode acoustic analysis and deep learning. The method comprises the following steps: acquiring an audio signal through an acoustic sensor, and extracting a mixed feature vector (including a Mel frequency spectrum and a time domain distortion parameter) by using an adaptive pre-emphasis and dynamic framing technology; the cascade noise reduction module separates noise from equipment abnormal sound, and a residual attention mechanism enhances abnormal features; a health model is constructed based on an improved LSTM-FCN network, and a decision boundary is dynamically optimized in combination with online incremental learning, so that non-intrusive anomaly detection and fault classification are realized. Compared with a traditional scheme, voiceprint recognition and equipment monitoring are fused, health assessment can be completed without an external instrument, the method has the advantages of being high in adaptability, low in false alarm rate, low in maintenance cost and the like, an efficient acoustic diagnosis scheme is provided for Internet of Things equipment, and application of edge intelligence in industrial predictive maintenance is promoted.

The invention discloses a data and model heterogeneous federated learning method of global decision boundary distillation learning. Two key problems of data heterogeneity and heterogeneous model performance difference are respectively solved through two sub-methods. According to the first sub-method, aiming at the problem of data heterogeneity, multiple prototypes are locally clustered, private data field distribution information is fully condensed, irrelevant distillation is introduced in a local supervised learning stage, and local knowledge forgetting is relieved. According to the second sub-method, global decision boundary distillation learning is provided for the problem of performance difference of the heterogeneous model, a global decision boundary learner is maintained and updated in a server, the local model is optimized from the perspective of the global decision boundary, and the influence of conflict information on the local model is reduced. The performance of the method is superior to that of various most advanced federal learning methods, the method can better adapt to challenging scenes, and meanwhile, the method has higher communication efficiency and better privacy reliability.

The invention relates to the technical field of fault diagnosis, and particularly discloses an electric spindle fault diagnosis method and system based on multi-scale dynamic convolution, and the method comprises the steps: collecting a multi-channel vibration signal of an electric spindle through a vibration sensor, carrying out the preprocessing, and segmenting the multi-channel vibration signal into equal-length samples through a sliding time window; constructing a multi-scale dynamic convolution module, extracting local and global features of an input sample through a parallel multi-branch structure, and dynamically aggregating convolution kernel weights of different scales based on an attention mechanism; the extracted local and global features are input into an AMSoftmax module, and decision boundaries between fault categories are increased through angle interval constraint; and training a diagnosis model by adopting a combined loss function, adjusting hyper-parameters in combination with Bayesian optimization, and outputting a composite fault classification result. Local and global features contained in a vibration signal are captured through parallel convolution kernels of different scales, an expert mechanism is introduced to realize adaptive selection of convolution kernel parameters, and the adaptability of a model to composite fault mode changes is enhanced.

The invention provides a large language model generation code detection method and system, which is applied to the technical field of artificial intelligence, and comprises the following steps: obtaining a to-be-detected code; a to-be-detected code is input to a trained shared encoder, a code feature vector is obtained, the shared encoder is obtained through multi-target joint training, and the multi-target joint training is used for optimizing classification loss, domain confrontation loss, comparison loss and difficult sample loss at the same time; l2 normalization is carried out on the code feature vector, and the code feature vector is mapped to a hyperspherical space to obtain spherical embedding; the sphere is embedded and input into a sphere category classifier based on sphere logistic regression for classification processing, a detection result of the to-be-detected code output by the sphere category classifier is obtained, the detection result comprises AI generation and human writing, and a decision boundary of the sphere category classifier is an intersection line of a hyperplane and a hypersphere for classification. According to the invention, the AI generation code and the human compiled code can be accurately distinguished.

The invention discloses a gesture recognition method and device based on a space-time decoupling graph convolutional network, and the method comprises the steps: generating inversion sample data through a teacher model obtained through old task data training, fusing the inversion sample data with current task data, and forming a fusion data set containing the knowledge of an old task and the features of a new task, incremental updating is carried out on a student model by adopting a domain consistency regularization loss function and a category boundary separation regularization loss function on a fusion data set, so that the problem of feature drift caused by difference between new and old sample domains is relieved, and a clear new and old category decision boundary is constructed; therefore, the disastrous forgetting phenomenon is effectively prevented, so that when the gesture recognition task is carried out in the dynamic environment, rapid adaptation of the current student model to new data is ensured, stable storage of knowledge of old tasks is kept, and the accuracy of gesture recognition in the dynamic scene is improved.