107 results about "Time series classification" patented technology

The invention discloses a supply chain intelligent tracking system oriented to process monitoring, relates to the technical field of supply chain intelligent management, and is used for solving the problems of flexibility and anti-risk capability of supply chain breakpoint management. Comprising a supply chain data processing module, an energy transmission analysis module, a breakpoint prediction module and a supply chain adjustment module. According to the method, original data are collected from supply chain nodes, effective features are generated in combination with a time sequence, classification features and a high-dimensional data processing technology, a breakpoint prediction model is constructed through a gradient lifting decision tree, breakpoint risks are identified and evaluated, a supply chain network topology structure is constructed, and propagation paths and influence ranges of the breakpoint risks are calculated recursively. Determining an affected node set, formulating an adjustment and recovery strategy in combination with a prediction result, optimizing resource allocation and logistics routes, quickly recovering key nodes, ensuring continuity of a supply chain, and upgrading an early warning mechanism to dynamically monitor risks, thereby improving management flexibility and anti-risk capability of the supply chain.

The invention relates to the technical field of multi-modal time series data processing, in particular to a multi-scale multi-modal fusion time series classification model, which comprises the following steps of: performing timestamp unification, missing value filling and normalization processing on input multi-modal time series data to generate standardized data; extracting long-period and short-period features by adopting a time sequence convolutional network and a one-dimensional convolutional network respectively, and performing feature alignment and fusion; cross-modal dynamic coupling is realized through cross attention and a gating mechanism, and feature noise reduction is performed in combination with adaptive threshold filtering; feature weighted fusion is completed based on multi-level feature division and a channel attention mechanism, and a comprehensive time sequence feature vector is generated; and finally, through nonlinear feature enhancement, Softmax probability prediction, sliding window smoothing processing and majority voting, outputting a classification result aligned with a timestamp. The method has the advantages of being high in feature fine granularity, good in cross-modal fusion effect and the like, and is suitable for the fields of intelligent manufacturing, behavior recognition, medical monitoring and the like.

The present disclosure relates to systems, non-transitory computer-readable media, and methods for detecting and validating anomalies from an ongoing data collection by applying an algorithm corresponding to a data pattern of the ongoing data collection. In particular, in one or more embodiments, the disclosed systems utilize a time series classification model to identify a data pattern corresponding to the ongoing data collection. Further, the disclosed systems can utilize an algorithm corresponding to the data pattern to monitor the ongoing data collection for anomaly candidates. Additionally, in one or more embodiments, the disclosed systems pass anomaly candidates through an anomaly validation filter to remove false positives.

The invention belongs to the technical field of data mining and artificial intelligence, and discloses a boxing action recognition method and system based on priori knowledge and multivariate time sequence classification, a non-contact action capture system is adopted to obtain skeleton point motion time sequence data, and a skeleton structure is stabilized through coarse-grained filtering based on unscented Kalman filtering, so that the recognition accuracy of the boxing action is improved. Fine-grained filtering based on kinematics priori knowledge inhibits noise and abnormity on a time sequence coordinate value level, comprehensive analysis is carried out on skeleton point motion time sequence data, and attack and defense actions in a boxing scene are effectively identified and classified by utilizing the kinematics priori knowledge and a multivariable time sequence classifier. According to the method, the problem that the action recognition accuracy is reduced due to low data quality caused by shielding is effectively solved, and a real-time feedback result is provided for a boxing scene based on the multivariable time sequence classifier of the dual-channel Transform.

The invention discloses a multivariate time sequence classification method fusing a convolutional neural network and Transform, and the method comprises the steps: constructing a multivariate time sequence classification model which comprises a multi-scale convolution module, a channel fusion convolution module, a feature dimension multi-head attention module and a time dimension multi-head attention module; fusion features are obtained through a multi-scale convolution module and a channel fusion convolution module, feature time dimension abstract features are obtained based on a feature dimension multi-head attention module and a time dimension multi-head attention module, and a prediction category is further obtained. According to the method, the local and global features of the time sequence can be extracted at the same time, and the limitation of the traditional CNN and Transform models during independent processing of the local or global features is overcome. Rich local features are extracted through multi-scale convolution operation, fusion of cross-channel information is enhanced through a channel fusion mechanism, and expressive force of the model in complex time sequence data is improved.

The invention provides a forest degeneration process identification and degeneration degree division method based on a time sequence, and relates to the field of forest resource degeneration restoration. The method comprises the following steps: acquiring surface reflectance data, and calculating an NBR index after preprocessing; fitting an NBR time sequence through a LandTrendr algorithm, and extracting sample place time sequence data based on random sampling and visual interpretation; generating an adversarial network and expanding sample point fitting data into a simulation data set in combination with a self-supervised learning technology; training CNN, random forest and BOSSVS time sequence classification models, and constructing a hybrid classification model through an integrated voting strategy; and identifying a forest degeneration process by using a hybrid classification model, and generating a degeneration degree diagram through reclassification. The method can achieve the efficient recognition of the complex degradation process of the forest region, has good time sequence adaptability and regional applicability, and provides technical support for the monitoring and management of ecological restoration of regional forests.

Monitors having housings, gas-specific sensors, environmental sensors, volatile organic compound sensors, indicators, processors, and memory configured to store event data and trained time-series classification models. Methods of monitoring comprising training a time-series classification model, directing environmental air towards sensors, receiving sensor data, determining event activity using the time-series classification model and providing the determination to an indicator.

The invention discloses a multivariable time sequence classification method based on a depth map neural network. The method comprises the following steps: 1, constructing a multivariable time sequence sample relation graph; 2, extracting potential spatial features of the MTS by using a variational auto-encoder; 3, extracting MTS sample relation features by using a depth map convolutional network, and fusing the MTS sample relation features with potential spatial features; and 4, calculating a loss function and iteratively updating the model so as to obtain an optimal variable time sequence classification model. In combination with multivariable time sequence classification, a deep neural network and a depth generation model, a sample structure information extractor based on a depth map neural network is provided so as to deeply mine structure information between multivariable time sequence samples; and a depth generation model is introduced to extract low-dimensional potential spatial features, so that the problem of over-smoothing of the depth map convolutional network when the number of layers is increased is solved, and relatively high classification accuracy can be achieved on the multi-variable time sequence classification problem.

The invention discloses an electric vehicle time sequence classification aggregation frequency modulation optimization model considering responsivity. The model comprises the following steps: establishing a power energy operation boundary for performing energy exchange of an EV; classifying each single EV according to the EV network access time and the network access time period, and aggregating the power operation boundary and the energy operation boundary of the single EVs with the same network access time period to obtain an EVA time sequence classification aggregation model; depicting the relationship between different types of EV responsivity and price excitation through the Weber-Fechner law according to the EV classification to obtain an EVA time sequence classification aggregation model considering the responsivity under charging and discharging excitation; and by considering the day-ahead and real-time two-stage income and cost of the EV participating in the energy-frequency modulation market after EV classification aggregation, and by taking the maximum EVA operation income as a target, establishing an EVA time sequence classification aggregation frequency modulation optimization model capable of giving consideration to both precision and efficiency. The model is used for accurately and efficiently evaluating the EVA frequency modulation capability, formulating an incentive mechanism and optimizing an EVA operation strategy.

The invention discloses a multi-label time sequence classification method based on spatio-temporal characteristics and dynamic loss. The method comprises three core modules including a data preprocessing module, a spatio-temporal characteristic learning module and a multi-task optimization module. The data preprocessing module adopts filtering and standardization processing to improve the input quality; the spatio-temporal feature learning module firstly constructs a multi-scale packet convolutional network to extract the time sequence feature of each channel, then adaptively adjusts the channel weight by constructing a channel calibration mechanism to enhance key information, and then constructs a cascade feature multiplexing network to extract spatial correlation features to realize deep fusion of spatio-temporal features, so as to improve the time sequence feature of each channel. And finally, the extracted spatial-temporal features are used for classification. The multi-task optimization module establishes a dynamic loss proportion weighted multi-task learning strategy, updates historical loss through an index moving average mechanism, and solves the problem of class imbalance through loss inverse proportion weight. Experimental results show that the classification accuracy of the method on the multi-label time sequence data set is remarkably improved, and the method has high practicability.

The invention discloses a bacterial colony image time sequence classification tracking method based on dynamic time warping and label propagation, and solves the problems of inconsistent classification and accurate segmentation of bacterial colonies with different sizes caused by the change of bacterial colony morphology along with time. The method comprises the following steps: acquiring a bacterial colony image feature sequence and constructing a bacterial colony feature distance matrix; calculating an accumulated distance matrix based on a dynamic time warping algorithm, and calculating accumulated distance values point by point through a recursion formula; through a path backtracking algorithm, finding a time point matching path which minimizes an accumulated distance value between bacterial colonies, and obtaining an optimal time alignment relationship between feature sequences of different bacterial colonies; constructing a similarity matrix, and calculating a similarity value between bacterial colonies; and a classification corresponding relation between time points is established by adopting a cross-time-point target association algorithm, and the bacterial colony classification label at the first time point is propagated to the subsequent time point, so that the consistency of the time sequence classification labels is ensured.

The invention discloses a demand optimization-oriented hydrothermal geothermal resource evaluation method, and relates to the technical field of geothermal resource quality and development and utilization time sequence classification evaluation, and the method comprises the steps: determining a relative importance relationship between evaluation indexes of an evaluation object through an expert scoring system, and carrying out the evaluation of the evaluation object based on the relative importance relationship. And constructing a judgment matrix for the evaluation object, determining the weight and the standardized numerical value of each evaluation index, multiplying and accumulating the weight of each evaluation index and the corresponding standardized numerical value to obtain a comprehensive score, and determining an evaluation result of the evaluation object according to the comprehensive score. It can be seen that the relative importance relation of the evaluation indexes is determined through the expert scoring system, the indexes of the resource demand side and the resource supply side are comprehensively considered, the judgment matrix is constructed, accurate evaluation of the evaluation object is achieved, and therefore the actual development decision is better guided.

The invention discloses a Fisher-Rao distance-based normal cloud similarity time sequence classification method, which relates to the technical field of time sequence classification, and comprises a sequence preprocessing module, a cloud concept generation module and a similarity measurement and classification module, and comprises the following steps: S1, the preprocessing module preprocesses input time sequence data, performing segmentation processing on the original time sequence and the difference sequence according to a target dimension; s2, converting each segmented sequence into a cloud concept containing an expectation Ex, an entropy En and a hyper-entropy He by using a reverse cloud algorithm in a cloud concept generation module; s3, constructing an index structure in a similarity measurement and classification module, sorting expectations Ex of segmented cloud concepts of training set samples in an original sequence and a differential sequence, and establishing a search structural body comprising sorted Ex values and corresponding original position indexes; s4, classifying by adopting a three-layer screening mechanism; according to the method, the classification performance is ensured, and meanwhile, the calculation efficiency is high.

A method classifies missing labels. The method computes, using a neural network model trained on training data, rank-based statistics of a feature of a time series segment to attempt to select two candidate labels from the training data that the segment most likely belongs to. The method classifies the segment using k-NN-based classification applied to the training data, responsive to the two candidate labels being present in the training data. The method classifies the segment by hypothesis testing, responsive to only one candidate label being present in the training data. The method classifies the segment into a class with higher values of the rank-based statistics from among a plurality of classes with different values of the rank-based statistics, responsive to no candidate labels being present in the training data. The method corrects a prediction by an applicable one of the classifying steps by majority voting with time windows.

The invention discloses a saliency-guided time sequence adversarial sample generation method and a saliency-guided time sequence adversarial sample generation system, and the method comprises the steps: obtaining a target time sequence classification model and a corresponding original input sample, carrying out the supervised training of the model through a time sequence training set, so as to guarantee the prediction accuracy of the model for the original input sample, and setting a disturbance iteration parameter; generating a saliency map of the input sample based on the target time sequence classification model; constructing a composite loss function fusing classification loss and saliency alignment loss for the target time sequence classification model based on the saliency map, carrying out iterative optimization on the model based on a projection gradient descent method framework and in combination with a saliency guidance strategy to update disturbance, terminating the optimization process to obtain updated disturbance when an iteration condition is reached, and carrying out the optimization of the target time sequence classification model. And outputting a final confrontation sample based on the obtained disturbance, and completing the generation of the confrontation sample of the time sequence. The attack success rate and the non-concealment of the time sequence confrontation sample are considered at the same time.

The invention belongs to the technical field of computers, and particularly relates to a multivariate time sequence classification method based on multi-scale decomposition, a program, equipment and a storage medium. According to the method, a multi-scale time slice method is introduced, an input time sequence is divided into non-overlapping time slices along a time dimension, modeling is performed on subsequences in different layers, and by using different time slice sizes in different PatchMama layers, a model can capture multi-scale features in time sequence data, so that the classification accuracy is improved, and the classification efficiency is improved. And the model can process data of different time scales, and has relatively strong generalization ability. The PatchMama model designed by the invention fuses two perspectives of different time resolutions and time distances, and can more accurately capture complex structures and dynamic changes in time sequence data, so that information in the data is more comprehensively understood, and the classification capability of MTSC problems is remarkably improved.

The invention relates to a multivariable time sequence classification method based on a bidirectional time and frequency domain state space model. According to the method, global information and frequency characteristics in multivariable time sequence data are effectively extracted by integrating a bidirectional time sequence and a frequency domain state space model, so that the classification accuracy and robustness are improved. The method comprises the following specific steps: carrying out mark coding and position coding on input data; inputting the coded data into a multi-layer time frequency domain Mama module, extracting global and frequency domain features, and applying layer normalization; performing nonlinear activation and Dropout operation on the output data, and projecting the output data to a classification dimension through a linear layer after the output data is flattened; and finally, outputting prediction distribution and carrying out end-to-end training by minimizing cross entropy loss. When the method is used for processing a multivariable time sequence classification task, the model performance and the calculation efficiency are remarkably improved.

The invention provides an industrial fault classification method, system and device and a readable storage medium, and the method comprises the steps: carrying out the serialization of fault data, carrying out the modeling of the data through a TimeGAN-based time sequence generation algorithm, generating the time sequence data of a fault, and achieving the expansion of the fault data. And finally, performing classification modeling on the expanded data by using a time sequence classification algorithm to realize fault classification. According to the method, the accuracy and robustness of fault classification can be improved, the problem of classification performance reduction caused by data scarcity is solved, and the adaptive capacity of the model in a complex and changeable industrial environment is enhanced. Besides, through a data generation mode, a training data set can be enriched, effective learning of a deep learning model is promoted, and then the intelligent level of industrial fault monitoring and management is improved.

The invention provides a steam temperature autonomous cruise method based on time sequence classification, which comprises the following steps of: acquiring historical data and preprocessing the historical data to establish a Transform model; collecting current data for preprocessing, and inputting the trained model to obtain a steam temperature state classification result so as to identify an operation state; when the identification result is the abnormal working condition, a preset expert strategy library is called to match a corresponding control strategy, and whether a progressive steam temperature set value compensation strategy is additionally executed or not is detected in real time until the abnormal working condition is finished; meanwhile, the model is updated and the expert strategy library is corrected according to historical correction records. By establishing the time sequence classification model, working conditions under different trends can be classified carefully and effectively, a control strategy is correspondingly selected for correction, and a set value compensation strategy is timely combined under the condition that expectation is not met, so that the response speed and the regulation and control precision of the system can be improved; and a guarantee is provided for safe and stable operation of the thermal generator set.

The application discloses a behavior intention open set recognition method and system based on a neural network and electronic equipment, and belongs to the field of intention recognition. The method firstly adopts a known class classification network comprising a self-attention hybrid convolution module and a time sequence change perception module to respectively extract key local features and global time sequence changes in the time dimension, and process the same into low-dimensional class activation vectors; then, in the training process, an adaptive threshold detection module is used to fit Weibull distribution of each action intention class and an unknown action intention class judgment threshold, and in the test or reasoning stage, the class activation vectors of test samples are corrected, and samples less than the threshold of the corresponding class are identified as the unknown action intention class. Action intention open set recognition experiments on various time sequence classification data sets show that the overall performance of the method is better than that of each baseline method, and ablation experiments of the model prove the rationality of the network structure.

The invention discloses a time sequence classification method and system based on a multi-relation atlas, and the method comprises the steps: obtaining multivariable time sequence data through the sampling of a sensor, and defining the state of the sensor; related symptoms of each sensor state are mined through a large language model, and a sensor state relation graph is constructed for each symptom; a multi-graph gating network is constructed and trained, the multi-graph gating network is composed of a plurality of GNNs, a gating network and an MLP network, each GNN is learned by using a sensor state relation graph of one symptom, multivariable time sequence data is learned through the plurality of GNNs to obtain time sequence characteristics of the sensor state relation graph of each symptom, and the time sequence characteristics of the sensor state relation graph of each symptom are obtained; and aggregating into a total symptom feature through a gating network, and finally obtaining final classification output through an MLP network. The invention relates to the technical field of natural language processing, can combine a multi-relationship graph with multivariable time sequence data training, enables the construction of the relationship graph to be flexible, and effectively improves the classification accuracy.

This invention discloses a lightweight adaptive optimization method for high-dimensional hyperparameters of convolutional neural networks (CNNs) for non-stationary time-series signal classification. It aims to address the technical challenges of performance degradation in time-series signal classification models and the reliance on expensive real-world evaluations for hyperparameter configuration under non-stationary perturbation scenarios. This method uses a deep convolutional neural network as the core classification carrier, treating the hyperparameter combinations within the deep convolutional neural network as decision variables to be optimized. With robust classification error rate, computational complexity, and training time as core optimization objectives, it constructs a closed-loop collaborative optimization mechanism of "perception-evaluation-decision" and utilizes a meta-learning dual-branch convolutional polynomial surrogate-assisted evolutionary algorithm (MetaDCP-SAEA) to achieve efficient configuration. This method requires no manual intervention; the convolutional neural network used for classifying non-stationary time-series signals can automatically search for the optimal hyperparameter combination. In simulated non-stationary noise environments, the reduction in classification accuracy can be controlled within 9.17%. It is suitable for robust classification scenarios of non-stationary time-series signals such as industrial IoT monitoring and medical signal diagnosis. It helps to lower the engineering threshold of artificial intelligence technology, promotes the large-scale application of automatic machine learning in complex environments, and has broad market prospects and application value.

The invention relates to the technical field of data processing, and provides a time sequence classification method and device, electronic equipment and a storage medium, and the method comprises the following steps: segmenting to-be-classified time sequence data according to a preset sliding window size, and forming a plurality of local time periods; for each local time period, processing the local time period by using a convolutional neural network so as to extract a feature vector of the local time period; inputting each extracted feature vector into a full connection layer, and generating a corresponding sparse response vector through the transformation of a weight matrix and the nonlinear effect of an activation function; performing frequency statistics on all the generated sparse response vectors, and constructing a characteristic frequency histogram of the time sequence based on a statistical result; and using the characteristic frequency histogram as representation of a time sequence, and inputting the characteristic frequency histogram into a classifier for classification so as to obtain a category label of the time sequence.

The application discloses an exoplanet light variation signal classification method based on transfer learning, acquires light variation signals of two telescopes for detecting exoplanets by using the transit method, and respectively uses the light variation signals as a pre-training model using dataset 1 and a target model using dataset 2; the acquired data is preprocessed; the two dataset data are respectively divided into a training set and a test set; a convolutional neural network model is constructed by using the dataset 1 to perform training, an optimal time series data classification model is established to perform time series classification; the pre-training model is saved; the dataset 2 is used to adjust the model, and a model suitable for the dataset 2 is obtained. The application avoids the limitation that a model cannot be learned under the condition of a small sample, and accelerates intelligent data exoplanet detection.

This invention discloses a method and system for fault diagnosis of industrial equipment based on a multi-view generation algorithm, belonging to the field of machine learning technology, to solve the problem of low fault diagnosis accuracy caused by inaccurate feature extraction in existing fault diagnosis methods. The key technical points of this invention include: collecting one or more operating state variables of industrial equipment, and sorting the same operating state variables by time to form an original univariate time series set; generating multiple views from multiple time series samples, and combining time series transformation with feature vector concatenation to generate a feature vector set for multiple views; inputting the feature vector of each view into a machine learning-based fault diagnosis classifier for training; generating multiple views from the time series of the operating state sample to be tested to obtain the corresponding feature vector; and inputting the feature vector into the trained classifier to obtain the fault prediction result. This invention significantly improves the accuracy and performance of time series classification and fault diagnosis.

The invention relates to an aperiodic time sequence classification method based on multi-view wavelet convolution and frequency domain enhancement and a computer readable storage medium, and the method comprises the steps: carrying out the multi-scale decomposition and equilibrium factorization of an input aperiodic time sequence, and mapping the aperiodic time sequence into a two-dimensional matrix; performing two-dimensional continuous wavelet transform on each matrix, and extracting an approximate component, a horizontal component and a vertical component; inputting the components into a multi-view wavelet convolution module, extracting local features by using different convolution kernels, and performing weighted fusion with SE channel attention through a gating mechanism to obtain multi-scale local features; performing FFT (Fast Fourier Transform) on the original sequence, and reconstructing global trend characteristics through iFFT after low-pass filtering and linear mapping enhancement; and finally, fusing local and global features, and outputting a result through a classifier. The method solves the problem of insufficient feature extraction of non-periodic and multi-stage time series data, improves the classification precision, has the advantages of small parameter quantity and high reasoning speed, and is suitable for real-time quality monitoring of industrial edge equipment.

The application provides a time sequence classification method, a classification terminal and a storage medium. The method comprises the following steps: preprocessing initial time sequence data, and extracting features of the preprocessed initial time sequence data by using an FCN network to obtain initial features; extracting the initial features by using an IAM network and an AGRes2net network to obtain global features; and classifying based on the global features to obtain a target classification result. The IAM network and the AGRes2net network are combined, multi-scale features can be extracted, and the classification result is more accurate.

The invention discloses a time-space spectrum hierarchical coupling time sequence classification method and system based on edge prior guidance. The method comprises the steps of obtaining a satellite image time sequence, and extracting spectral band information, a normalized vegetation index and coordinate information of each pixel; time features are obtained through the normalized vegetation index; coding the coordinate information to obtain spatial features; spectral gradient collaborative characteristics are obtained through spectral band information; performing filtering processing on the satellite image time sequence to generate a binary edge mask; constructing a space-time spectrum feature mining model to obtain space-time spectrum comprehensive features; and adjusting the channel dimension of the space-time spectrum comprehensive feature so as to be matched with the number of the classification categories. According to the method, a space-time spectrum feature mining model is constructed, the advantages of a residual network and a self-attention mechanism can be reserved, and it is ensured that the model focuses on multi-level local features. Therefore, multi-scale space-time spectrum comprehensive features can be extracted from the enhanced feature representation, and the crop classification capability is improved.

A method for assessing and / or monitoring a process and / or a multi-axis machine includes recording at least one data time series, wherein the at least one data time series includes at least one channel describing at least one parameter of the process and / or of the multi-axis machine, and wherein the data time series is caused by the process. An interpretable result is determined by a machine learning algorithm based on the at least one data time series, wherein the result describes a classification value of a state in the process and / or of a state of the multi-axis machine. A warning is output when determining the result if the classification value of the state in the process and / or of the state of the multi-axis machine is assigned to a value of an error class that is in a warning range or corresponds to a warning range, and an all-clear signal is output if the classification value of the state in the process and / or of the state of the multi-axis machine is assigned to a value of an error class that is in an all-clear range or corresponds to an all-clear range.

The invention provides a video-based identification model for Parkinson's disease and idiopathic tremor. The video-based identification model comprises an attitude estimation module, a data processing module and a classification module, the attitude estimation module comprises a pre-trained RTMPose-L model, performs whole body attitude estimation on three upper limb motion videos of a subject frame by frame, extracts key point coordinates of a wrist and five fingers of a target hand, and forms an absolute coordinate sequence of multiple groups of hand key points; the data processing module converts the absolute coordinate sequence into a relative coordinate sequence and calculates a statistical characteristic sequence of the speed, the acceleration, the amplitude, the frequency and the entropy; the classification module comprises a PatchTST time sequence classification model, the relative coordinate sequence and the statistical feature sequence are divided into a plurality of patches with preset lengths, then feature extraction is carried out on each patch through a Transform encoder, patch prediction is output, finally, an output result is flattened into a one-dimensional vector, and the one-dimensional vector is subjected to patch prediction. And the Parkinson's disease and the idiopathic tremor are classified through the full connection layer.