694 results about "Time series dataset" patented technology

The invention relates to an electric power material intelligent detection method based on multi-modal data fusion, and aims to improve the accuracy and automation level of material state recognition. According to the method, in the electric power material operation or circulation process, multi-modal data such as images, infrared thermal imaging, radio frequency identification, vibration response and environmental parameters are acquired through a unified time index, and a structured time sequence data set is constructed. And after normalization and exception elimination processing, multi-dimensional feature vectors including structural strength, temperature distribution, label continuity and dynamic stability are extracted, and weighted statistics and correlation calculation are executed to generate a comprehensive state index. And further through comparison with a historical reference, identifying an abnormal state according to a deviation threshold value, outputting corresponding labels and feature information, and obtaining material state evaluation and disposal suggestions based on rule reasoning. According to the method, accurate monitoring and abnormal early warning of the electric power materials under the driving of the multi-source data are realized, and the method has good practicability and expansibility.

The invention discloses a hydropower station equipment fault analysis method based on state data mining, and relates to the technical field of hydropower station equipment intelligent fault diagnosis, and the method comprises the steps: collecting key operation parameters through deploying multiple types of sensors, and constructing a unified state time series data set; carrying out supervised training by adopting an LSTM network, extracting a dynamic feature vector, and constructing an AI state analysis model; introducing a micro-fluctuation abnormal coefficient WBYX, and evaluating the operation stability of the equipment; a coupling disturbance collaboration coefficient OHRD is calculated, and a fault conduction relation among multiple devices is identified; and calculating a trend evolution coefficient QSYH based on the state vector included angle offset, and analyzing whether the equipment operation trend is abnormal or not. By setting a multi-level threshold value, generation of a hierarchical early warning mechanism and a response strategy is realized, and the operation safety and the fault prediction capability of hydropower station equipment are effectively improved. The method is suitable for hydropower station key equipment state monitoring and intelligent operation and maintenance management in a complex environment.

The present disclosure relates to a multi-factor coupling cooperative early warning method and system for fatigue crack propagation of a steel structure. The multi-factor coupling cooperative early warning method includes: obtaining multi-physical field monitoring data of a dangerous source distribution point of a steel structure project, and obtaining a monitoring time series data set; establishing an intuitionistic fuzzy matrix of the monitoring time series data set; obtaining uncertainties of indexes by using a grey relation coefficient between the monitoring indexes of physical fields; taking obtained uncertainties as basic probability assignments of evidences; preprocessing the evidences through weighed averaging, and obtaining corrected basic probability assignments; obtaining basic probability assignments of fatigue crack propagation of the steel structure at different development stages; and determining a fatigue crack propagation grade of the dangerous source distribution point of the steel structure project through the basis probability assignment.

The provided are a price elasticity analysis and prediction method and model based on deep learning. The model consists of a CNN layer and an RNN layer. The method comprises the following steps: S1, collecting historical data and merging the historical data into a multi-dimensional time series dataset; S2, extracting sentiment data and trend data from market news and social media; S3, inputting the data obtained into CNN for data preprocessing and feature extraction; S4, inputting the feature extracted by CNN into RNN for time series analysis; S5, training and optimizing model: using Adam algorithm to adjust the learning rate adaptively, and combining the momentum method and RMSProp algorithm to improve the generalization ability and prediction accuracy of the model. The provided combines the advantages of CNN and RNN, which can understand and predict the complex relationship between price and market behavior more comprehensively and accurately.

A network traffic classification process, including the steps of: monitoring network traffic flows to dynamically generate, for each of the network traffic flows and in real-time, time series data sets representing, for each of upstream and downstream directions of the network traffic flow, for each of a plurality of successive timeslots, and for each of a plurality of packet length bins, a packet count and a byte count of packets received within the timeslot and having one or more lengths within the corresponding packet length bin; and processing the time series data sets of each network traffic flow to classify the network flow into one of a plurality of predetermined network traffic classes, without using payload content of the network traffic flow.

The invention discloses an operation collaborative optimization method for an optical storage direct current flexible interaction system. Comprising the steps of collecting operation data such as photovoltaic output, an energy storage state, household load power and direct current bus transmission power, fusing power market price information, and constructing a multi-dimensional time series data set; then, predicting an adjustable load capacity interval of the system based on a coupled physical constraint neural network model embedded with DC bus power balance, voltage constraint and equipment operation limitation; further constructing a state-action space, solving a Pareto frontier by adopting a multi-objective optimization algorithm, and generating a light storage and home load collaborative scheduling strategy set; then combining the real-time operation state and the prediction deviation information, applying a voltage-power droop control mechanism to carry out strategy decoupling, and generating an energy storage power correction amount and a flexible load priority control instruction; and finally, a control instruction is issued to the optical storage direct flexible system, so that collaborative optimization operation with consideration of economical efficiency, safety and comfort of the system is realized.

The invention relates to the technical field of automobile part manufacturing, and discloses an automobile injection molding part production process control system and method, and the system comprises the following modules: a data collection module which is used for collecting technological parameters, molds, raw materials and equipment operation original data, attaching timestamps, and storing the data in a database; the process parameter prediction module is used for reading original data to construct a time sequence data set, inputting the time sequence data set into a TFT model to obtain a pre-training model, and predicting a short-term process parameter fluctuation range in combination with current production working condition parameters; and the quality risk index acquisition module is used for inputting the process parameter data and the mold data into a quality risk index calculation formula to obtain a quality risk index. Through the system, data-driven comprehensive production optimization is realized, the process control accuracy and adaptability are improved, the quality control scientificity and reliability are enhanced, the intelligent level of the production process is improved, the production efficiency is effectively improved, the defective rate is reduced, and the product quality is stabilized. The problem that process control lacks system intelligence is solved.

The invention discloses a cognitive impairment man-machine interaction method and system based on emotion analysis, and relates to the field of cognitive impairment man-machine interaction, and the method comprises the steps: collecting the physiological information, language information and expression information of a user in real time, carrying out the feature fusion, forming a time series data set, and carrying out the emotion recognition and classification; based on a user emotion recognition classification result, establishing a generative adversarial network to generate man-machine interaction dialogue content matched with the current context and emotion in real time; when it is detected that the emotion recognition result of the user is negative emotion, a cognitive disorder emotion pacifying strategy is retrieved based on big data, emotion pacifying is conducted on the user, and related data are recorded; by analyzing historical related data of negative emotions of a user, an emotion prediction model is constructed, possible emotion changes are recognized in advance, and a man-machine interaction pacifying strategy is adjusted.

The invention discloses an industrial internet real-time cooperative control method based on edge computing, and the method comprises the steps: obtaining multi-source heterogeneous data of industrial equipment, and generating a standardized data set; performing time sequence alignment processing on the standardized data set to generate a time sequence data set; segmenting the data to obtain a standardized segmented data set; removing abnormal values and redundant information according to the standardized segmented data set to obtain a cleaned data set; performing format conversion on the cleaned data set according to an equipment state and a control strategy requirement to generate intermediate data; performing feature extraction on the intermediate data to generate an equipment state feature vector; according to real-time monitoring information of a network load and a transmission path, a computing resource allocation strategy of an edge node is adjusted, data of high-priority equipment is processed preferentially, and a resource allocation result is generated; and deploying a cooperative control algorithm based on an ARIMA time sequence model at the edge node, and generating a real-time control instruction in combination with the equipment state feature vector.

The invention discloses an industrial time sequence prediction method based on adaptive continuous learning. The method comprises the following steps: firstly, dividing a non-stationary industrial time series data set to obtain a plurality of domains with the maximum distribution difference; different time domains are then modeled in sequence, and an improved empirical playback (DER + +) method is used to avoid catastrophic forgetting of previously accumulated knowledge. Meanwhile, a soft sample buffer area is introduced to promote memory and learning of key modes in the current field. And finally, the time-sensitive activation function TimeRelu enables the time convolutional network (TCN) to have a time evolution property, and the generalization ability of the prediction model is enhanced. According to the method, the continuous learning normal form is introduced into the time sequence prediction task, the limitations of huge resource overhead of traditional cumulative training, disastrous forgetting of an incremental learning mode and the like are overcome, and the method has theoretical and practical significance on industrial time sequence prediction.

The invention discloses a method for performing wind power generation prediction through a fine-tuning pre-training large model, and the method comprises the steps: obtaining a historical multi-dimensional time series data set, and carrying out the spatial-temporal feature fusion through a sliding window, and generating a wind speed trend sequence; based on the wind speed gradients of the adjacent time windows, generating step length adjustment parameters by using a normalization function; performing dynamic parameter adjustment on the pre-trained time sequence prediction model based on the step length adjustment parameters, and performing domain adaptation on top network parameters of the time sequence prediction model through an adaptive optimization algorithm in combination with a hierarchical transfer learning strategy to obtain an optimized time sequence prediction model; and processing the real-time multi-dimensional time sequence data set by using the optimized model to generate a wind power generation prediction result. By dynamically adjusting model parameters and a hierarchical transfer learning strategy, the precision of wind power generation prediction and the model adaptation capability are effectively improved, and the method is suitable for time sequence prediction scenes in the field of wind power generation.

Aiming at the problems of insufficient data diversity, data sparsity and model generalization ability in an industrial time sequence prediction model, the invention provides a method for establishing a time sequence prediction model oriented to industrial multi-modal data, which comprises the following steps: acquiring multi-source heterogeneous industrial time sequence data, and cleaning the acquired data; performing data enhancement on the cleaned data to generate an enhanced time sequence data set; fusing the multi-source heterogeneous time series data in the obtained enhanced time series data set by using a heterogeneous data fusion sensing method of multi-dimensional space-time mapping; using the fused data to construct a prediction model oriented to the multi-modal industrial time series data based on an AI large model, and performing model training; and according to the trained prediction model oriented to the multi-modal industrial time series data, predicting the change trend of the key indexes in the industrial process in a period of time in the future.

The invention discloses a regional power grid wind power generation power prediction optimization method and system, and particularly relates to the technical field of power grid wind power generation. Multi-source environment data is collected, wind speed change characteristics are extracted, a time series data set is constructed, a hybrid prediction model is constructed in combination with a long-short term memory network and a gradient boosting decision tree, a time series trend and a nonlinear wind speed-power mapping relation are captured respectively, and prediction weights of the time series trend and the nonlinear wind speed-power mapping relation are adaptively adjusted through an attention mechanism. According to the method, the model hyper-parameters are adaptively adjusted by further combining Bayesian optimization, short-term prediction errors are corrected by using Kalman filtering, the real-time adaptability and stability of prediction are enhanced, the problems of prediction misalignment and the like caused by high nonlinearity and data scarcity of a wind speed mode in a complex terrain environment are effectively solved, and the prediction accuracy is improved. The wind power prediction precision is improved, a more stable and reliable scheduling reference is provided for a regional power grid, and the standby capacity demand and the operation cost are reduced.

The invention relates to a line loss data anomaly monitoring method and system based on time sequence anomaly detection. The system comprises the following steps: S1, collecting operation data of a power grid; s2, preprocessing the acquired operation data of the power grid; s3, according to the time sequence data set, through a Patch Transform, segmenting a time sequence into local windows and calculating attention, and through GraphSAGE, aggregating line topology correlation features to obtain a training feature set; s4, constructing a hierarchical anomaly detection model and training based on the training feature set; and S5, performing anomaly detection according to the trained layered anomaly detection model, and generating a causal thermodynamic diagram based on an anomaly detection result. According to the invention, through multi-source data fusion, dynamic distribution modeling and causal analysis, accurate identification and diagnosis of abnormal reasons in a complex line loss scene are realized.

The invention discloses a river water level dynamic monitoring and flood overflow risk prediction method based on deep learning, and the method comprises the following steps: S1, collecting multi-source hydrological data, and constructing a time series data set; s2, performing interpolation, denoising and normalization processing on the data to generate a unified time sequence format; s3, constructing a water level prediction model comprising a bidirectional long short-term memory network and an attention mechanism; s4, inputting the preprocessed data into the water level prediction model, and outputting a multi-time-step predicted water level sequence; s5, a dynamic threshold value is set according to the historical extreme value and the real-time hydrological condition, and the flood overflow risk is judged; s6, generating and caching a risk tag, and recording an error; s7, outputting a prediction result and risk information through a communication interface; and S8, periodically updating the input data in a rolling manner, and repeatedly executing the prediction and monitoring process. According to the invention, depth prediction and a dynamic threshold control mechanism are fused, and water level monitoring and flood overflow risk intelligent early warning are realized.

The invention relates to the technical field of slurry valve design, and discloses a large-diameter slurry valve sealing structure design method and device suitable for a high-wear working condition, and the method comprises the steps: firstly obtaining key geometric parameters of a sealing structure, constructing a three-dimensional topological basic model and a mechanical property simulation model, and then carrying out the fusion to form a digital twin model; then constructing a wear morphology time sequence data set; meanwhile, three-level early warning threshold values are set for the key working condition parameters, a working condition risk grade division rule is established, and a real-time working condition risk grade is obtained according to the working condition risk grade division rule; and finally, based on the wear morphology time sequence data set and the real-time working condition risk level, constructing a wear evolution prediction model, and optimizing key geometric parameters of the sealing structure in combination with a digital twin model to obtain an optimal geometric parameter combination. Accurate prediction and optimal design of the abrasion state of the sealing structure of the large-diameter slurry valve can be achieved, the service life of the sealing structure under the high-abrasion working condition is remarkably prolonged, and the reliability of the sealing structure under the high-abrasion working condition is remarkably improved.

The invention provides a cross-time-space conversion method with a Fourier filtering operator for multivariable time sequence classification. The cross-time-space conversion method comprises the following steps: mapping an input time sequence to a target variable; mapping a time sequence data set to a group of C category labels Y; after training, the CSTFormer model outputs a vector containing C values so as to estimate the probability that the time sequence belongs to each category; in the CSTFormer model, data input by a multivariable time sequence firstly passes through a LayerNorm layer and then enters a global adaptive Fourier neural operator module, noise is adaptively filtered, and important frequency information is enhanced; the output passes through a LayerNorm layer and then enters a time-space consistency learning module; the time-space consistency learning module is composed of three sub-modules, namely a dynamic multi-scale time attention module, a dynamic space attention module and a dynamic cross-distribution fusion module; respectively learning a time context and a space context, and fusing to obtain a learned time-space consistency context; and finally, outputting a classification result by the classification head.

The invention discloses a single-person abnormal behavior identification method and system based on multi-modal skeleton feature fusion, and the method comprises the steps: S1, collecting continuous RGB images and infrared thermal imaging images in a monitoring video, carrying out the human body detection and key point estimation of visible light and infrared images through employing a multi-modal fusion model of YOLOv12 in combination with Transform, and constructing a single-person posture time series data set; s2, key point speed vectors are calculated for the continuous skeleton frame sequence of each target person, skeleton key point information and speed information are fused, and an action feature sequence is formed; s3, inputting the motion feature sequence into an MPED-RNN model, decomposing skeleton motion into a global displacement component and a local attitude deformation component, and performing joint coding, decoding and prediction through a dual-channel GRU network; and S4, calculating a prediction error and a reconstruction error according to a reconstruction result and a future skeleton key point prediction result, evaluating whether the current behavior deviates from a normal trajectory, and judging whether the current behavior is in an abnormal state. According to the invention, real-time identification of abnormal behaviors of a single person in a complex scene is realized.

Systems and methods of anomaly detection in time-series are disclosed. A time-series dataset is received and a set of segments is iteratively defined from the time-series dataset by identifying a set of changepoints in the time-series dataset based on a changepoint type and a sensitivity parameter, determining whether the set of segments defined by the changepoints satisfy at least one threshold criteria, and modifying the sensitivity parameter when the threshold criteria is not met or outputting the set of segments when the threshold criteria is met. A segment-specific anomaly detection threshold is determined for each segment in the set of segments and a set of anomaly-flagged segments is generated. The set of anomaly flagged segments are generated by an anomaly detection process based on the segment-specific anomaly detection threshold for a corresponding segment. An anomaly-flagged time-series is generated by combining the set of anomaly-flagged segments.

The invention provides a vehicle equipment consumption prediction method and device, a model training method and device, a computer readable storage medium and electronic equipment, and relates to the technical field of vehicle equipment maintenance prediction, and the method comprises the steps: obtaining the vehicle equipment time sequence data of to-be-predicted vehicle equipment; according to the vehicle equipment time sequence data, through a Transform-KAN model, obtaining a prediction result of vehicle equipment consumption; wherein the Transform-KAN model is obtained by training an improved Transform model on the basis of a vehicle equipment time sequence data set, and the improved Transform model is obtained by replacing an MLP layer in the Transform model with a KAN model. According to the invention, the technical effect of rapidly and effectively detecting the vehicle equipment is realized.

The invention discloses an intelligent networked battery full life cycle optimization system and method, and relates to the technical field of energy management, the system comprises a sensor array module, temperature data and voltage data in the battery operation process are continuously collected through a sensor array, an initial multi-dimensional time sequence data set is formed, and the initial multi-dimensional time sequence data set is stored in the sensor array module; the alarm and adaptive adjustment module is used for obtaining an alarm log record aiming at the initial time sequence data set, triggering the adaptive adjustment module to dynamically optimize battery operation parameters according to the alarm log record, generating a temporary control instruction, reducing the influence range of temperature or voltage abnormity and obtaining adjusted operation state data; according to the intelligent networked battery full life cycle optimization system and method, risk monitoring of the battery full life cycle is realized, and the battery operation safety and reliability are effectively improved.

The invention relates to the technical field of big data and the field of financial science and technology, and discloses a decision generation method based on big data, and the method comprises the steps: collecting multi-dimensional data related to a target business, extracting key features, such as a behavior mode, a potential problem and resource stability, of a customer, constructing a customer analysis model, and carrying out the decision generation. Generating a time sequence data set in combination with the behavior trend data; a behavior data set is generated through scene analysis and quantitative analysis is carried out to obtain a behavior mode and a resource utilization trend of a client, and finally a strategy adapted to each business scene is generated. According to the method, the service strategy is finally generated by collecting the multi-dimensional data, extracting the customer characteristics, constructing the analysis model, analyzing the customer behavior trend, generating the time series data set and combining different service scenes to perform scene analysis and quantitative processing, so that the real-time analysis, scene decision and dynamic adjustment of the multi-dimensional data are realized; the flexibility and accuracy of the system in a complex service scene are improved, and the service decision-making efficiency and accuracy are improved.

The invention belongs to the field of multivariate time series data missing value filling, and discloses a time series missing value filling method and system based on a diffusion model, and the method comprises the steps: 1, dividing a time series data set into a training set, a verification set and a test set, and carrying out the data processing of each piece of time series data, obtaining a 01 mask matrix of each piece of time sequence data, a to-be-filled target and observable time sequence data; 2, a training process is carried out, and a trained denoising network is obtained; and step 3, inputting a forward noise adding result corresponding to the test set into the trained denoising network obtained in the step 2, and outputting a reverse denoising result as filled time sequence data. According to the method, the flexibility of noise weight setting is realized, and the applicability and universality of the model to different time sequence data sets are enhanced.

Computerized methodologies are disclosed that are directed to detecting anomalies within a time-series data set. An aspect of the anomaly detection process includes determining one or more seasonality patterns that correspond to a specific time-series data set by evaluating a set of candidate seasonality patterns (e.g., hourly, daily, weekly, day-start off-sets, etc.). The evaluation of a candidate seasonality pattern may include dividing the time-series data set into a collection of subsequences based on the particular candidate seasonality pattern. Further, the collection of subsequences may be divided into clusters and a silhouette score may be computed to measure the clustering quality of the candidate seasonality pattern. In some instances, the candidate seasonality pattern having the highest silhouette score is selected and utilized in anomaly detection process. In other instances, a plurality of seasonality patterns may be combined forming a time policy, where the time policy is utilized in anomaly detection process.

A method for selecting forecasting models for generating timeseries workload forecasts for a contact center covering varying timeseries granularities and operating horizons. The method includes selecting, via a first selection process, a first select forecasting model from first candidate forecasting models for forecasting a workload level in accordance with a lower-granularity timeseries. The first selection process may include the steps of: receiving a first timeseries dataset; defining different timeseries datasets within the first timeseries dataset, including a first shorterm dataset and first longterm dataset; testing, using the first longterm dataset, each first candidate forecasting model in accordance with a first longterm cross-validation process; testing, using the first shorterm dataset, each first candidate forecasting model in accordance with a first shorterm cross-validation process; and calculating, for each of the first candidate forecasting models, an combined accuracy score based on the testing.

The invention relates to an uncertainty online evaluation method based on time sequence data decomposition and a Bayesian principle, and aims to solve the problem that a traditional uncertainty evaluation method is difficult to process online and dynamically update. The method comprises the following steps: firstly, collecting measurement data within a certain time range, constructing a time sequence data set, and decomposing the time sequence data into a trend period item and a random item through an STL decomposition technology; then, on the basis of the data of the random item, parameters of prior distribution and a likelihood function are calculated by using a Bayesian statistical method, joint posterior distribution is further determined, and the uncertainty of a measurement result is calculated on the basis of the posterior distribution; afterwards, the posterior distribution result at the current moment is used as the prior information of uncertainty evaluation at the next moment to be fused and updated with the new observation data, so that the real-time online evaluation of the measurement uncertainty is realized. The method not only improves the accuracy and real-time performance of the evaluation result, but also realizes the continuous updating and optimization of the measurement information, and is suitable for a complex and rapid-change measurement environment.

A method for a time series forecaster including a machine learning (ML) model and a time series forecasting algorithm may be used to generate a future forecast model for a time series dataset. The future forecast model includes a future prediction showing a future forecast of the time series dataset. Rather than simply presenting the importance of a forecasting model feature in a limited manner, such as in the form of a number (e.g., a float number), the future forecast model of one or more embodiments may advantageously include at least one forecasting model feature visualization output that visually illustrates the effects of one or more forecasting model features on the future prediction.

The invention discloses a forest fire risk assessment method combining microwave remote sensing and fire spreading simulation, and the method comprises the following steps: integrating microwave remote sensing data, and selecting a driving factor which affects the occurrence probability of a forest fire; establishing a time series data set based on the selected driving factor, and constructing a forest fire occurrence probability prediction model based on deep learning; simulating a fire spreading process through a cellular automaton-Monte Carlo simulation method, and calculating a potential combustion probability; and comprehensively evaluating the forest fire risk by combining the fire occurrence probability and the potential combustion probability. The forest fire risk comprehensive assessment method provided by the invention has the advantages of high accuracy, strong predictability and the like, the model can capture trends and changes of factors such as meteorology and vegetation within a period of time before a fire occurs, important features in driving factors can be better concerned, and the risk of the forest fire can be comprehensively assessed. The method for comprehensively evaluating the forest fire risk is more accurate in identification of high-risk fire areas, and more scientific and accurate information is provided for forest fire management.

The embodiment of the invention provides a water quality index prediction model training method and device, equipment and a storage medium, and the method comprises the steps: carrying out the data enhancement of a training data set through employing a meta-learner network after the training data set is obtained, so as to obtain an enhanced time series data set. And inputting the enhanced time sequence data set into a trained model, and calculating training loss through the trained model, thereby optimizing model parameters of the trained model according to the training loss. According to the method, data enhancement can be carried out by utilizing a meta-learner network, layered contrast learning is carried out on multiple dimensions of global, instance and time, and information of a time sequence on different semantic hierarchies is comprehensively captured. The sequence features are extracted by using the convolutional neural network, and the self-supervised hierarchical comparison model is trained and established, so that the accuracy of the water quality index prediction result can be improved, and the stability is relatively high.

The invention discloses a photovoltaic output time sequence data set generation method and system, and the method comprises the steps: carrying out the data preprocessing and simulation parameter setting, carrying out the transfer sampling of daily weather conditions in a simulation duration through a first-order Markov chain, introducing a local cloud layer influence model to carry out the all-day irradiance simulation of different daily weather, and carrying out the simulation of the irradiance. And calculating daily temperature based on the all-day simulated irradiance, calculating daily output power of the photovoltaic system by considering the influence of the daily temperature and the all-day simulated irradiance on the generating capacity, and screening out a photovoltaic time sequence data set meeting the daily generating capacity and the total generating capacity by adopting a double screening method. Appropriate time granularity and time scale can be selected according to needs, then the influence of meteorological changes on photovoltaic system output is rapidly simulated, a data set approaching the actual situation is generated, and the method can be applied to the fields of photovoltaic system performance evaluation, machine learning, optical storage integrated management and control and the like.