63 results about "Time dependency" patented technology

The invention relates to a long-range task-oriented VLA model method for a humanoid robot, which comprises the following steps of: S01, analyzing a natural language instruction through a space-time semantic analyzer to generate an atomic operation sequence with a space-time dependency relationship; s02, maintaining a task state machine by using a dynamic memory network, and tracking the task execution progress in real time; s03, integrating vision, language and sensor data through a multi-modal perception fusion engine; s04, calling a predefined action primitive based on an adaptive execution system and optimizing a motion track; and S05, performing online updating and optimization on the model through a continuous learning mechanism. According to the method, a natural language instruction is analyzed into a structured task sequence with space-time dependence through a space-time semantic analyzer, an execution sequence and preconditions are defined, the semantic understanding ability and the structuring degree of task planning are improved, and task decomposition and replanning in a dynamic environment are supported; according to the method, the LSTM and the knowledge graph are combined, and the task state is maintained in real time.

The invention discloses a multi-modal abnormal data detection and restoration method and system oriented to a power business scene. The method comprises the following steps: collecting multi-source heterogeneous power data, abstracting a power system into a weighted undirected graph, uniformly mapping the multi-source heterogeneous data into a graph signal, and preprocessing the collected data; extracting spatial features of nodes in a topological structure by adopting a graph convolutional network, and capturing a time dependency relationship in combination with a time sequence encoder; identifying various types of data abnormal points through an abnormal scoring function fusing the time sequence prediction error and the neighborhood consistency; a prediction-reconstruction combined repair strategy is adopted, time sequence prediction and neighborhood diffusion estimation are fused, and a preliminary repair value is generated; a lightweight parameter adapter is introduced, a scene feature vector is used as input, a repair weight and a regularization coefficient are dynamically generated, and a repair strategy is automatically adjusted; and performing physical consistency verification on a data result, wherein the physical consistency verification comprises power injection conservation constraint, voltage amplitude range constraint and time sequence continuity constraint.

The invention discloses a universal multi-modal target tracking method based on space-time propagation and modal cooperation, and belongs to the technical field of computer vision. The method comprises the following steps: converting RGB and X modal images into a token form, and constructing initial features in combination with modal specific time tokens; the method comprises the following steps of: extracting a multi-level enhanced feature through a Transform encoder and a Mama collaborative prompt block; generating discriminative fusion features by using a gating fusion and context sensing module; a time-guided attention mechanism is adopted to strengthen search area features, and a result is output through a tracking prediction head; and transmitting the fusion time token as historical information to the next frame, and dynamically updating the template by combining a long-short time template updating strategy. According to the method, complementarity and space-time dependence between modes are effectively mined, tracking robustness and generalization ability in a complex scene are improved, and the method is suitable for various mode combination tasks such as RGB-D, RGB-T and RGB-E.

The invention relates to the technical field of artificial intelligence, can be applied to business scenes such as financial science and technology and medical health, and discloses a root cause analysis method and device based on a space-time dependency graph, equipment and a medium. Comprising the steps of constructing a space-time dependency graph, generating a diagnosis path blueprint, identifying a fault source and a root cause entity type, generating a graph query statement, executing query and performing cause and effect verification, and outputting a root cause analysis report. And the causal reasoning and root cause positioning of the system state change are realized by fusing the graph structure information and the natural language processing capability. Through cooperative processing of a language model and a graph data structure, fault symptom information and system structured state data are deeply fused, a path is generated in the graph structure, and a causal relationship is verified, so that the ability of understanding a complex system state evolution chain is improved, and accurate identification and diagnosis of root causes are realized. And the accuracy and the automation level of root cause analysis are obviously enhanced.

The invention relates to the technical field of mechanical equipment intelligent operation and maintenance and state monitoring, and discloses a health monitoring method for bearing state evaluation and residual life and degradation trend prediction, which comprises the following steps of: firstly, extracting time domain, frequency domain and time-frequency domain characteristics from original vibration signals of a bearing under different working conditions; according to comprehensive evaluation indexes and known bearing degradation characteristics, features having good characterization capability and trend consistency for bearing degradation performance are screened out, a novel backbone network model is constructed, deep features are mined, effective features are enhanced, meanwhile, a time dependency relationship in a long sequence is captured, and then a multi-task learning mechanism is introduced, so that the bearing degradation performance is evaluated. Through parameter sharing and joint optimization, bearing state identification, residual life prediction and performance degradation trend prediction can be synchronously completed only by training and deploying a single model. According to the method, multi-task collaborative prediction under complex working conditions is realized, and the accuracy and robustness of bearing state recognition and service life prediction are improved.

The invention relates to the technical field of task scheduling, in particular to an intelligent routing inspection scheduling method and system for a construction site environment, and the method comprises the following steps: extracting construction layout equipment distribution and task types based on construction site environment information, identifying task space distribution and marking priorities, analyzing time constraint, evaluating conflicts, and adjusting task distribution. Extracting core nodes, analyzing connection strength, screening and classifying, calculating path coverage efficiency ranking priorities, screening efficient paths, and obtaining a construction site intelligent inspection scheduling table. According to the invention, by analyzing the spatial distribution, priority and time dependency of inspection tasks, optimizing task identification and organization, optimizing an execution sequence based on accessibility and spatial layout, dynamically adjusting an allocation scheme, improving path flexibility and cooperation efficiency, extracting key nodes to associate and optimize path logic, and screening efficient paths for scheduling; the resource utilization rate is improved, and the problems of insufficient coverage of regular inspection and fixed paths, scheduling stiffness and the like are solved.

The invention belongs to the technical field of computer networks, and discloses a resource-constrained network equipment-oriented double-layer GRU lightweight real-time network anomaly detection method, which comprises the following steps of: constructing an anomaly detection model and training, deploying the trained model to resource-constrained network equipment, the method comprises the following steps: acquiring a system log generated when network flow generated by each application server node flows through in real time, performing data preprocessing on the acquired system log to obtain log semantic vectors, and forming a window sequence by the continuous log semantic vectors to obtain an abnormal probability value; and comparing the abnormal probability value with a set threshold value, and directly outputting an output detection result at an equipment end. According to the method, complexity of entity information is eliminated, time dependence in a time sequence is enhanced, simplification is carried out, a high anomaly detection recall rate is achieved under the condition that low calculation overhead is kept, the risk of missing detection is reduced, and cascade faults caused by the fact that anomalies are not found in time are avoided.

The invention discloses a micro-service fault diagnosis method and system based on multi-modal data spatio-temporal dependency perception, and the method comprises the steps: fusing various modal data, such as logs, indexes and calls, through a gating index linear unit; and extracting time dependence and space dependence characteristics among multi-modal data through a UGFormer algorithm fusing a Transform attention mechanism and GNN graph structure perception capability, and realizing nonlinear transformation mapping from a characteristic vector to target output by using MLP so as to complete complex fault detection and root cause positioning tasks. Experimental evaluation results prove that in a representative micro-service fault data set, the accuracy rate of the method based on the multi-mode space-time dependency perception in a fault detection task reaches 98.9%, and the accuracy rate of the method based on the multi-mode space-time dependency perception in a root cause positioning task HRat1 (Top-1 hit rate) reaches 82.6%.

This invention discloses a rubidium atomic clock frequency adaptive prediction method and system based on LSTM network, belonging to the field of time and frequency technology. The invention includes the following steps: Step 1. Data acquisition; Step 2. Filtering and denoising preprocessing; Step 3. Normalization processing; Step 4. LSTM model training and deployment; Step 5. Frequency prediction; Step 6. Online model adaptation. This invention employs a Long Short-Term Memory (LSTM) network, a special type of recurrent neural network, which can effectively learn the complex time dependencies and noise patterns in rubidium atomic clock time and frequency data. The inherent gating mechanism of the LSTM model makes it adept at capturing long-term dependencies in time series, thus enabling high-precision prediction of short-term frequency fluctuations. Through an online fine-tuning mechanism, this invention allows the LSTM model to continuously adapt to the individual drift and aging characteristics of a single rubidium atomic clock, achieving precise optimization for each clock and model, significantly improving the practicality and long-term stability of the method.

The invention discloses an optical cable damage monitoring method, system and device based on GAT and LSTM algorithms, and a medium, and relates to the technical field of optical fiber communication monitoring, and the method comprises the steps: collecting an original signal, carrying out the noise reduction processing, carrying out the dynamic boundary segmentation of time series data after noise reduction, forming a time segment sequence, calculating the similarity between time segments through an elastic alignment mode, and carrying out the detection of the optical cable damage. The method comprises the following steps: constructing a graph structure, performing spatial feature extraction on the graph structure, dynamically distributing importance weights among nodes, generating feature vectors, inputting the feature vectors into a time sequence modeling unit, capturing time dependence, outputting final time sequence features, and performing disaster type classification and position positioning based on the final time sequence features. According to the method, the spatial-temporal characteristics are modeled cooperatively through the GAT-LSTM mixed architecture, the OTDR data are analyzed by using the graph structure, and the decoupling capability and the detection sensitivity of the optical cable composite disaster spatial-temporal propagation path are improved.

The invention discloses a central heating and refrigerating system power adjusting method, terminal equipment and a storage medium. All available time data are mapped into a hidden space. Data points located on the manifold between the points associated with the available training time data are selected so that new data can be generated that complies with a new time dependency. And the enhanced data is input into the graph structure network, so that the model prediction performance is improved. And lagging relationship representation is obtained by performing lagging relationship calculation on the time series data, future values of the time series data are predicted in an auxiliary manner, and the capturing capability of the model on the large inertia characteristic of the system is enhanced. The improved ST-GNN prediction model and an MPC optimization controller are deeply integrated to form a complete closed loop from perception, prediction, optimization, execution and feedback. A model online updating mechanism is introduced, when the prediction deviation is increased due to the change of system characteristics, the model can be quickly adjusted based on a small amount of new data, and the model performance is prevented from attenuating along with time.

The invention relates to the crossing field of medical data processing and artificial intelligence technology, in particular to an Alzheimer's disease progress prediction method and system fused with time dependency, and the method comprises the steps: obtaining a standardized data set, each sample in the standardized data set comprising data of at least three time points; constructing a dynamic multi-task time accumulation model, aiming at each sample, enabling data of each time point to correspond to one task, and taking the weighted sum of task prediction results of all time points as a model prediction value; a cross entropy loss function is adopted to calculate an error between a model prediction value and a real diagnosis label, and back propagation is carried out to update parameters; inputting to-be-tested data into the trained dynamic multi-task time accumulation model, and outputting prediction results of different time points; the method can achieve the precise prediction of the progress track of the whole course from normal cognitive impairment and mild cognitive impairment to definite diagnosis of Alzheimer's disease, and provides data support for clinical early intervention and personalized treatment scheme formulation.

The application discloses a kind of telemetry data prediction method, device, medium and product, it is related to spacecraft telemetry data prediction field.The method includes: obtaining the telemetry data corresponding to different variables of target spacecraft in each time in the time sequence to be measured, to obtain the real-time telemetry data sequence corresponding to each variable;Real-time telemetry data sequence corresponding to multiple variables is all input into trained multi-scale time convolution network, to obtain real-time time correlation feature;According to the real-time telemetry data sequence corresponding to multiple variables, based on similarity principle and trained multi-head attention mechanism, determine real-time graph structure feature;Real-time time correlation feature and real-time graph structure feature are all input into trained graph attention network, to obtain the telemetry prediction data corresponding to different variables of next time of current time.The application can simultaneously capture the time dependence of telemetry data and the spatial dependence between each variable, to further improve the prediction accuracy of spacecraft telemetry data.

The invention discloses an electrocorticogram motion trail prediction method based on a deep attention network, and belongs to the field of motor nerve decoding in a brain-computer interface. According to the method, the feature extraction and trajectory prediction performance of the ECoG electroencephalogram signals is improved. For the problem that an ECoG electroencephalogram signal has long-range time dependence, an ABL mechanism is introduced, local features of time dimension are extracted by using CNN in a feature extraction stage, and weight fusion is performed by using loss output by the local features and output loss of a main model, so that long-range features and short-time local features are concerned while long-range features are concerned; meanwhile, a Transform model is used for extracting global features in the feature extraction process; according to the method, the accuracy and robustness of model joint trajectory decoding are improved, and the method has remarkable advantages in the aspects of difficulty in effectively capturing a global time sequence mode, difficulty in capturing a complex mapping relation between an ECoG signal and a joint and the like.

The invention discloses a time action positioning method and system based on time sequence context maximum pooling. The method comprises the following steps: acquiring a to-be-identified action video, and extracting and acquiring a feature coding sequence of the to-be-identified action video; presetting a time action positioning model, inputting the feature coding sequence into the time action positioning model, and obtaining an action classification result; by introducing the time sequence context maximum pooling operation and the multi-scale time feature pyramid structure, the calculation complexity is effectively reduced, and the reasoning speed and efficiency are improved. Meanwhile, by designing a time action positioning model comprising an encoder, a long-term time context module and a decoder, the model can simultaneously capture short-time and long-time dependency, so that the accuracy of time action positioning is improved. Due to the advantages, the method has important application value in the field of time action positioning.

The invention provides an encrypted traffic classification method and device, equipment, a storage medium and a program product, and the method comprises the steps: obtaining a plurality of classification rule sets, and dividing an encrypted traffic set into first classification difficulty traffic and second classification difficulty traffic; determining a classification result of the first classification difficulty traffic; generating a session image based on the second classification difficulty traffic; decomposing the session image into a plurality of data packets, performing embedding processing, and capturing interaction information among all the data packets to obtain global features; extracting spatial features of the data packet, and extracting time dependence of the spatial features to obtain spatial-temporal features of the data packet; and fusing the global features and the spatial-temporal features to obtain fused features, and generating a classification result of the second classification difficulty traffic. Wherein the classification efficiency is improved by classifying and analyzing the encrypted traffic with different classification difficulties, and the classification precision is improved by classifying the encrypted traffic with higher classification difficulty on the basis of global features and spatio-temporal features of the encrypted traffic.

The invention relates to the technical field of flood disaster prediction, and discloses a flood disaster cascade event evolution prediction method and device, equipment and a storage medium. According to the method provided by the invention, entity nodes and relation types in a drainage basin are constructed into a knowledge graph which is updated along with time, and flood disaster cascade event evolution is reflected in real time by utilizing a node attribute dynamic updating and relation edge weight dynamic adjusting mechanism; meanwhile, time features are introduced to be embedded into knowledge graph nodes and relation edges, and time dependence represented by the nodes is enhanced; and finally, reasoning the knowledge graph by adopting a relationship evolution graph convolutional neural network, aggregating information of neighbor nodes through the graph convolutional neural network by utilizing each relationship edge, and adjusting respective propagation intensity based on a relationship type, so that a result output by the graph convolutional network not only pays attention to a structural relationship between the nodes, but also pays attention to the information of the neighbor nodes. And the edge weight is adjusted according to the change of node attributes, so that the development venation of flood disasters is accurately simulated.

The invention discloses a user behavior risk prediction method and device, equipment and a medium. The method comprises the following steps: constructing a graph structure according to feature data of a plurality of users; fusing the feature of each node in the graph structure with the feature of the corresponding neighbor node to obtain the static node feature of the node; determining the dynamic node characteristics of each node according to the characteristics of each node in the graph structure at different time points; and determining a behavior risk prediction result of a user corresponding to each node according to a fusion feature of the static node feature and the dynamic node feature of each node in the graph structure. According to the method, the static node features representing the graph structure features and the dynamic node features representing the time dependence are fused, so that the complex relationship between the nodes is represented, the behavior change features at different time points are dynamically and adaptively extracted, and the accuracy of user behavior risk prediction is improved.

The invention discloses a bearing small sample fault diagnosis method based on cross prediction, belongs to the field of bearing fault detection, and provides a time-frequency cross prediction (TFCP) self-supervised learning method aiming at the problems that label samples are difficult to obtain and the manual labeling cost is high in bearing intelligent fault diagnosis. According to the method, the dependence on manual labeling is reduced by extracting the deep features of the label-free fault signals. Firstly, a time-frequency data enhancement method is designed to provide support for a model to mine original signal features; secondly, developing a time-frequency cross prediction task training network, capturing a time-frequency sequence time dependency relationship and learning overall representation; and finally, optimizing parameters of the pre-training model by utilizing the 1% annotation data to realize fault identification. An evaluation result of a public data set CWRU and a PU bearing data set shows that the method can still show excellent and competitive performance by only using 1% of labeled data, and the application potential of the method in small sample fault diagnosis is verified.

The present disclosure relates to methods, systems, and apparatuses for featurizing time-series data to enhance machine learning model training. Time-series data, such as transaction records, is preprocessed to identify fields including descriptive and categorical information. Categories are assigned using a first machine learning model, and tags are applied based on domain-specific patterns or large language models. The processed data is organized into a star schema data structure comprising a fact structure and associated dimension structures. Features are generated from the data structure based on time windows, incorporating statistical metrics and identified patterns. These features are provided to a machine learning module to train a second machine learning model, improving accuracy and adaptability for applications such as customer behavior prediction and financial analysis. The disclosed approach addresses challenges of high dimensionality, noise, and temporal dependencies in time-series data, enabling robust and contextually relevant feature generation.

The invention provides a time sequence prediction model based on double domains and multiple variables, and the model simultaneously extracts a long-short time dependency relationship, an inter-variable dependency relationship and an intra-variable dependency relationship in a time domain and a frequency domain, so that the precision of time sequence prediction and the generalization ability of the model are improved. The model is composed of a time sequence input module, a decomposition embedding module, a trend prediction module, a remainder prediction module and a fusion output module. Firstly, mutual conversion of a time domain and a frequency domain is realized through DFT (Discrete Fourier Transform), secondly, a trend component and a remainder component are respectively predicted by adopting a double-branch structure, and finally, two prediction results are added to obtain a final prediction result of a multivariable long-time sequence. According to the invention, the average prediction precision of the model is improved; in the design of the model, the principle of modular design is followed, and a convolution architecture is adopted, so that the calculation complexity is effectively reduced, and the expandability and efficiency of the model are improved.

The invention belongs to the technical field of medical statistics and disease risk prediction, and discloses an organization state prediction model construction and verification method, device and system and a storage medium, and the method comprises the steps: dividing patient data into a derivation queue and an internal verification queue according to a certain proportion, and introducing at least two external verification queues; screening out at least one key variable from the plurality of initial variables by adopting minimum absolute shrinkage and selection operator regression; constructing a model through multivariable Cox proportional risk regression and presenting the model in a column diagram; performing multi-queue verification from distinguishing capability, calibration capability and risk layering; wherein the distinguishing ability comprises a Harley consistency index and an area under a 3 / 5 / 8 year time dependence curve, and the risk layering comprises a Karlan-Myer method and a logarithmic rank test. And the finally constructed model is accurate in prediction and strong in generalization ability, and can intuitively and efficiently assist in organization state risk assessment.

The invention relates to the technical field of network security, and discloses an XSS attack detection method and system based on multi-model fusion, and the system comprises a traffic data cleaning module, a feature extraction module, a traffic detection module and an instruction generation module. According to the method, hidden features in the front-back direction are captured by adopting a BiTCN to better obtain long-time dependency of a sequence, then, output of the BiTCN is further processed by utilizing a BiGRU layer, and the BiGRU enhances the memory ability of a model by combining a forward GRU and a reverse GRU, so that the model can learn dynamic changes of data from two directions, and therefore, the reliability of the model is improved. The method improves the perception capability of the model for the dynamic change of the time sequence, reinforces the semantic representation of the key segment of the XSS attack through the multi-head self-attention dynamic distribution weight, can reinforce the interaction between the internal features of the model through weight distribution and an attention layer, enables the model to learn more complex and abstract feature representation, and improves the accuracy of the model. And the problem of deep network gradient disappearance is solved by combining residual connection.

The invention discloses a bearing segmented life prediction method and system based on a graph neural network, and relates to the technical field of rotating machinery health monitoring. The method comprises the following steps: acquiring a vibration signal of a to-be-detected bearing; based on the health index of the bearing, dividing the whole life process of the bearing into two stages of a stable stage and a degeneration stage by using a life division model; the service life prediction model is used for conducting service life prediction on the vibration signals in the stable period and the degradation period respectively, prediction results are combined, a final bearing service life prediction result is obtained, the service life prediction model constructs the vibration signals into multi-mode time scale features, and the service life prediction model is used for predicting the service life of the bearing. And carrying out mining analysis on time dependence in the multi-modal time scale features so as to realize life prediction. According to the method, a two-stage division strategy fused with deep learning is combined with an efficient mining analysis method of time scale information, and efficient prediction of the residual service life of the bearing is achieved.

The invention discloses a secure transaction processing method and system, electronic equipment and a storage medium. The method comprises the steps of obtaining multiple pieces of transaction data of a user within a preset time period; extracting key feature information related to the transaction behavior from the transaction data; forming feature sequences by the key feature information of the plurality of transaction data; inputting the feature sequence into a pre-trained LSTM model, and judging whether transaction abnormity exists or not; if the transaction is abnormal, analyzing whether the plurality of transaction data are in a preset security control range; and if not, sending an abnormal message to remind a user. According to the method, the time dependence in the transaction data is captured by using the sequence data processing capability of the LSTM model, so that the accuracy of anomaly recognition is improved; meanwhile, when the transaction is identified to be abnormal, whether the transaction abnormal data is in a preset safety control range is evaluated, so that normal transaction fluctuation and potential malicious transaction behaviors can be distinguished more accurately and effectively.

Embodiments of the present application provide a transaction behavior identification method, device, equipment, storage medium and program product, relating to the fields of artificial intelligence and financial technology. The method comprises: acquiring time series data corresponding to a transaction behavior, the time series data comprising transaction feature vectors of a plurality of time steps; performing feature extraction on the transaction feature vectors through a self-attention mechanism to generate global features, the global features containing global dependency relationships; inputting the global features into a bidirectional time series modeling module to extract local time dependencies of the global features through forward and reverse sequence modeling to generate bidirectional hidden states; and performing classification processing on the bidirectional hidden states to obtain an abnormal probability of the transaction behavior being abnormal. The method of the present application significantly improves the identification capability of hidden abnormal transaction behaviors, solves the performance deficiency problem caused by single models in traditional methods, and achieves high-precision, low-false-alarm-rate abnormal transaction behavior detection effects.

The invention discloses a high-performance time sequence parallel simulation system and method based on a time acceleration and synchronous control mechanism, and belongs to the technical field of high-performance parallel calculation. The system comprises a timing trigger module, a time mapping engine module, a task generation and distribution module, a parallel processing module, a synchronous blocking module, a state aggregation module and a visual output module. By establishing a double-layer time system of physical time takt-simulation time window, a time mapping engine dynamically calculates the number of virtual time steps needing to be propelled according to acceleration multiplied speed, and acceleration mapping of physical time and simulation time is achieved. The parallel processing module executes tasks in a thread pool with a fixed size, and the synchronous blocking module performs unified time sequence control on execution of each thread through a counting synchronizer, so that state consistency and calculation reproducibility under a high-speed operation condition are ensured. And the state aggregation module is used for integrating the synchronized local calculation results and updating the global system state. According to the method, high-speed simulation and high-fidelity playback of large-scale time-dependent data are realized on the premise of not damaging sequential logic, and the modeling verification efficiency and stability of a complex system can be remarkably improved.

The invention discloses a shield tunneling composite stratum intelligent identification method and system based on a time sequence convolution space state model and a self-adaptive threshold value, and the method comprises the steps: collecting a real-time monitoring data set for shield tunneling composite stratum identification, carrying out the preprocessing of the real-time monitoring data set, and generating an initial stratum time sequence data feature; extracting and enhancing the initial stratum time sequence data features by using a state space model, and obtaining time sequence dynamic features; according to the invention, under the condition of keeping linear calculation complexity, joint optimization is carried out on a long-time dependency relationship and class imbalance, and high-precision and low-delay stratum type identification is carried out on a shield tunneling site; and the judgment threshold is dynamically updated by introducing an online feedback mechanism, so that the recall rate of a small number of rare categories and the accuracy of common categories can be considered, the recognition accuracy of tunneling stratum recognition is remarkably improved, the reliability of a tunneling stratum recognition result is guaranteed, and the method is suitable for being widely popularized and used.

The invention discloses a knowledge graph enhanced multivariate time series data anomaly detection method, which belongs to the field of data anomaly detection, and comprises the following steps of: decoupling by using two detection paths, namely a time path and a space path, through a Transform and variational auto-encoder model correlation algorithm, extracting the space-time dependency of complex multivariate time series data, and extracting the time-space dependency of the multivariate time series data; therefore, a database in industrial production, especially abnormal data in communication uplink throughput data, is discriminated and investigated, and the industrial operation efficiency and the communication data security are effectively improved. The method has higher accuracy for data with strong space-time coupling characteristics, is wide in universality, can accurately detect different types of industrial data, especially communication system data, by adopting a machine learning method, is strong in generalization ability and strong in real-time performance, can detect and dynamically update data abnormal conditions in real time, and can discover network problems in time.

The invention provides an electric power communication line insulation degradation early-stage anomaly detection method and a medium, relates to the field of anomaly detection, and provides an EWED anomaly detection model for long-sequence-dependent electric power communication line data, and the EWED anomaly detection model is composed of a long-time dependence module, a frequency spectrum coding module and an anomaly judgment module. The identification capability of the model for early abnormal trends is enhanced by the long-time dependence module; the frequency spectrum coding module enhances the resolution of insulation degradation early-stage features; the anomaly judgment module improves the accuracy and response sensitivity of anomaly judgment, the interpretability and controllability of results are enhanced through probabilistic expression, and the system has the capacity of being flexibly adjusted to adapt to different operation environments through a threshold mechanism.