181 results about "Time index" patented technology

The invention relates to a task scheduling optimization method and device based on reinforcement learning, equipment and a medium. The method comprises the steps that firstly, system resource state data are collected in real time, dynamic environment characteristics are determined through preprocessing and time sequence analysis, task characteristic data are analyzed at the same time, and a task priority sequence and a resource demand vector are generated through a priority ranking algorithm and a resource evaluation model; and then a state space and an action space are constructed by adopting a reinforcement learning algorithm, an optimal task allocation scheme is generated through strategy iteration and reward function optimization, and if the scheme meets a resource balance threshold, scheduling is executed, and performance indexes are collected. And finally, fusing real-time indexes with historical data, and updating parameters of the reinforcement learning model through experience playback and gradient descent to form a closed-loop optimized improved scheduling strategy. By adopting the method, the accurate mapping of the resource state and the task requirement can be realized, and the problem of insufficient adaptability of the traditional static scheduling to a complex scene is solved.

The invention discloses a micro-service dynamic weight load balancing method and related equipment. The method comprises the following steps: asynchronously acquiring a multi-dimensional real-time load index of a micro-service instance, obtaining a load prediction value by virtue of a model, fusing the real-time index, the prediction value and a service health degree, calculating a dynamic weight by depending on service load data, distributing a request according to the weight, and feeding back an adjustment coefficient. According to the method, by asynchronously collecting the multi-dimensional real-time indexes, an accurate basis is provided for weight calculation, so that the request allocates the actual load of the matched instance, and the situation that the response of a high-load instance is slow due to too many requests is avoided; by introducing a load prediction value, load change can be pre-judged in advance, a weight strategy is adjusted in advance, overload when an instance load suddenly increases is prevented, and the fluctuation resistance of the system is improved; the weight is calculated by fusing multi-dimensional data, so that the comprehensive load of the instance can be comprehensively reflected, and evaluation deviation caused by a single index is avoided; a feedback mechanism can enable weight calculation to continuously adapt to a system state, and load unevenness caused by weight fixing or adjustment lag is avoided.

The invention provides a data exception early warning method and device based on a dynamic threshold model. The method comprises the following steps: firstly, constructing an index time sequence library and an event library for storing index and event information; fusing the historical same-period base line, the trend term and the event influence value and generating a base line through a dynamic weight formula; calculating upper and lower limits of a dynamic threshold by combining the fluctuation acceleration and a basic fluctuation coefficient; and finally, comparing the real-time index value with a threshold range to judge whether to give an alarm, and optimizing the model or updating the event library according to a manual labeling result. According to the method, dynamic and accurate early warning of data exception can be realized, service scene changes can be adapted, the early warning accuracy is continuously improved through feedback optimization, and powerful support is provided for stable service operation and efficient exception disposal.

The invention discloses an intelligent master station self-healing control method and device and a computer program product, and the method comprises the steps: building a self-healing control evaluation index system which comprises a response time index, an accuracy index and a validity index; constructing a self-healing control model based on the index system, wherein the self-healing control model comprises an action judgment logic, a switch refusing judgment logic and a fault recovery constraint condition; a multi-objective ant lion optimization algorithm is improved, an initial population is generated through chaotic mapping, and algorithm parameters are dynamically adjusted; and solving the self-healing control model by using an improved algorithm, generating a Pareto frontier solution set through multi-objective optimization, and outputting an optimized fault isolation and power restoration control strategy. Through deep combination of algorithm optimization and model constraint, high precision, quick response and global optimization of self-healing control are realized, the reliability and anti-interference capability of a power system are remarkably improved, and continuous and safe operation of a power grid under a fault is ensured.

The invention relates to the technical field, in particular to an unmanned inspection method and system for a switch cabinet, and the method comprises the following steps: building an incidence relation between signal nodes through a graph neural network, improving the time sequence consistency of multi-source signals, enabling instrument reading and temperature and humidity signals to form a hierarchical structure under a unified time frame, and improving the accuracy of inspection. The logic association of state layering is enhanced, continuous sampling sequence features are extracted through a long-short-term memory network, the response sensitivity and time prediction capability to abnormal signal abrupt change are improved, amplitude matching of temperature difference and current signals is achieved under the same time index, early abnormal nodes are accurately recognized, and a dynamic positioning sequence is established. Through joint tradeoff of distance and energy consumption, path lengthens and energy peak values are reduced, task throughput and equipment availability are improved, timing sequence capture and trend judgment capabilities are enhanced in an anomaly recognition layer, and accuracy and continuity of inspection decision and reliability of safety assessment are integrally improved.

The invention relates to the field of data query, and discloses a data aggregation query method and system based on time dimensionalities, the method comprises the following steps: obtaining a data aggregation query request, the data aggregation query request at least comprising a time index table of the time dimensionalities, the time dimensionalities being represented by at least one of year, month, day, hour and minute; analyzing the data aggregation query request, and determining an aggregation granularity, a date index value and an aggregation step length in a time index table; performing aggregation calculation according to the aggregation granularity, the date index value and the aggregation step length, and determining an aggregation date index; and performing field matching in a database through the aggregation date index, and performing grouping aggregation on matching results to determine a data query result. Indexes are added for different aggregation dimensions by creating a date index table, and an index calculation algorithm is combined, so that multiple time dimension query requirements are met, the data query speed can be remarkably increased, consumption of calculation resources is reduced, and the query flexibility is improved.

The invention provides a remote crusher cluster intelligent scheduling method and system based on 5G communication, and relates to the technical field of wireless communication, and the method comprises the steps: collecting the real-time load, feeding intensity, yield, conveying equipment flow and other data of crushers through a 5G network; a fluid dynamic model containing upstream-downstream material transportation time delay parameters is adopted in the central dispatching server, and downstream material input is predicted in combination with the material attribute tags; and generating a target production rhythm curve with a time index, forming a time-related cooperative control instruction, and issuing the time-related cooperative control instruction to an equipment end controller for synchronous execution. According to the method, operation parameters are adjusted in a self-adaptive mode according to feeding strength differences of different cavity types, the yield and mechanical loads are balanced, and the instantaneous accumulation and overload risks are reduced; the execution precision and robustness under network jitter are improved through mechanisms such as time slot alignment, time delay compensation and sub-lattice point protection; and the productivity utilization rate is increased, equipment abrasion is reduced, and production continuity and safety are guaranteed.

The invention discloses a UUV trajectory data enhancement method, system and device based on time sequence fusion, and the method comprises the steps: obtaining the position data, time index data and navigation data of an underwater unmanned vehicle, and analyzing the position data to obtain three-dimensional distance data; performing normalization processing on the time index data and the three-dimensional distance data to obtain an initial time sequence data set, performing distance analysis on the initial time sequence data set to obtain initial distance data, and performing preprocessing to obtain a time sequence data set; establishing a time series data clustering model to analyze the time series data set to obtain a clustering time series data set; performing iterative updating through a particle swarm optimization algorithm and a fitness function to obtain an updated time sequence data set; and performing fitting correction on the updated time sequence data set to obtain a corrected time sequence data set. According to the method, the time index data is combined, updating is carried out through the particle swarm optimization algorithm, and then the accuracy of trajectory data simulation of the unmanned underwater vehicle is improved through fitting correction.

The embodiment of the invention provides a data storage method and a data query method, and the data storage method comprises the steps: obtaining to-be-stored data, and determining a target storage unit corresponding to the to-be-stored data; extracting at least one piece of storage time information from the to-be-stored data; at least one target storage fragment is determined according to the at least one piece of storage time information and fragment time indexes corresponding to a plurality of storage fragments in a target storage unit, and the storage time information is in one-to-one correspondence with the target storage fragments; and storing the to-be-stored data to the at least one target storage fragment according to the at least one piece of storage time information. The to-be-stored data is stored in the at least one target storage fragment in the target storage unit according to the storage time information, fragment storage of the data in different time periods is achieved, the data locality is better, the hit rate is higher, and due to the fact that the data popularity is strongly related to the time periods, data cold and hot separation is achieved, and the data storage efficiency is improved. And the combination overhead of cold and hot data is reduced.

The invention relates to a man-machine interaction user action intention recognition method, device and equipment and a medium, and the method comprises the steps: for continuous wave trough values of a wave trough value cluster, according to a first time index corresponding to a wave peak value in a wave peak value cluster and a second time index corresponding to the wave trough value in the wave trough value cluster, calculating the continuous wave trough values of the wave trough value cluster; finding out a middle crest value between the continuous trough values to segment a plurality of action segments in the gesture action; inputting the to-be-recognized gesture action into the gesture action recognition model to determine probability distribution corresponding to the gesture action category of the to-be-recognized gesture action; and adopting a preset Monte Carlo simulation algorithm to perform multiple random sampling on the probability distribution corresponding to the gesture action category, and taking the gesture action category with the maximum probability as the user action intention of the gesture action to be recognized so as to complete the recognition of the user action intention of human-computer interaction. According to the method and the device, smoother, more personalized and more immersive control experience can be provided for the user.

The invention provides a method and a device for quickly identifying an abnormal transaction event, and electronic equipment. The method comprises the steps that transaction data is acquired, the transaction data at least comprises an event identification code, a risk control index and a time window corresponding to the risk control index, the event identification code is a unique identification code representing a transaction event, the risk control index is an index used for determining the risk of the transaction event, and the time window represents the length of a time statistics window of the risk control index; dividing risk control indexes into short-time indexes and long-time indexes according to time windows in the transaction data; and calculating the short-time index, calculating the long-time index by adopting an Flink distributed stream processing framework to obtain a risk control index value, and carrying out transaction interception under the condition that the risk control index value is greater than a preset threshold value so as to avoid occurrence of a transaction abnormal event. Through the method and the device, the problem of user property loss caused by incapability of timely finding financial abnormal behaviors due to poor stability of a wind control calculation method in the prior art is solved.

The invention provides a computing power center computing power scheduling method based on a multi-dimensional weight algorithm and a storage medium, and the scheduling method comprises the following steps: S1, obtaining a submitted task of a user through a scheduling portal, forwarding the task to a scheduling management service, and obtaining metadata information such as a resource demand, a priority and deadline of the task by the scheduling management service; s2, the resource management service system obtains the newest NodeMetric of each computing node through a unified monitoring service, task types are divided according to the task resource demand characteristics, and the NodeMetric is a node real-time index; and S3, the scheduling management service system generates Wfinal according to a multi-dimensional index based on the task type through a weight calculator, and sorts the candidate nodes from high to low according to the Wfinal.

The invention discloses a regional meteorological model error control method based on a neural network, and particularly relates to the field of regional meteorological model error control, and the method comprises the steps: obtaining reanalysis data of a first data source and station observation data of a second data source, and forming an input matrix containing space coordinates, time indexes and meteorological element values; preprocessing operation is executed on the input matrix, and preprocessing comprises the steps that interpolation correction is conducted on missing values, normalization processing is conducted on data of different dimensions, sequence alignment is executed based on a unified time index, and a first input matrix and a second input matrix which are consistent in structure are formed. By constructing a double-residual channel and executing residual reciprocal correction in a training process, dynamic balance fusion of reanalysis data and station observation data is realized on a spatial-temporal scale, so that the problem of cross-source data fusion imbalance in the prior art is solved.

An audio description system includes a memory and a processor. The memory stores source media comprising frames positioned within the source media according to a time index. The processor is configured to generate, using an image-to-text model, a textual description of each frame; identify intervals within the time index, each interval encompassing one or more positions of one or more frames; identify placement periods within the time index, each placement period being temporally proximal to an interval; generate a summary description based on at least one textual description of at least one frame positioned within a selected interval temporally proximal to a placement period; and associate the summary description with the placement period.

The invention discloses a holographic communication system-oriented multi-dimensional interaction quality perception and resource optimization scheduling method, which comprises the following steps of: firstly, constructing a system containing objective performance and subjective experience indexes according to a scene, acquiring network, holographic and user layer data, training a subjective experience representation model, and estimating the subjective experience indexes in real time; estimating indexes at the current moment and the next moment, constructing a design vector, and fusing physical, semantic and cognitive layer multi-source knowledge to obtain an enhanced input vector; reasoning the dynamic weight of the index through a utility function and a fuzzy rule, and calculating an interaction quality comprehensive score; screening indexes to be optimized, converting the indexes to be optimized into a multi-objective optimization problem by taking a comprehensive score reaching the standard as a constraint, and searching a compromise optimal scheme; and finally, converting the scheme into a resource configuration scheme to realize self-adaptive scheduling of network and terminal resources. Accurate mapping of objective indexes and subjective experience is realized, scenes and user requirements are dynamically adapted, and immersion experience and communication performance of holographic communication are improved.

The invention relates to a gateway access threshold dynamic adjustment method and device based on Transform, and belongs to the technical field of information. Comprising the following steps: analyzing the influence condition of each data dimension on a host load, and selecting the dimension with large influence as a host load index; carrying out discretization, vectorization and position coding on the load index value sequence of the continuous time slice to convert the load index value sequence into an input vector; performing self-mask training on the input vector by using a Transform model to obtain a prediction model; predicting according to the real-time index value of the host by using a prediction model to obtain a predicted load index of the next time period; and comparing the predicted load index result with a preset load target value, and adjusting a flow load threshold value of the gateway host according to a comparison result. The invention mainly solves the problems that the threshold setting is not accurate enough and the fixed threshold is difficult to adapt to a rapidly changing service scene because the traditional gateway load threshold is a fixed pre-estimated value, and improves the utilization rate of host resources.

A method for generating soft information for a digital signal decoder. The method includes receiving a first plurality of signals, each corresponding to a time index, each time index being associated with two states including state 1 and state 0. The method also includes processing the first plurality of signals to calculate path metric values of state 0 and 1 for generating each of a second plurality of most-likely symbols by hard decisions and intermediate information for a most-likely path and a second-likely path at each time index. The method includes computing a difference between the most-likely path and a second-likely path for a given one of the two states at each time index. The method also includes determining a soft information corresponding to either one of the two states at the time index per one of the second plurality of most-likely symbols.

The invention provides a real-time incremental RAG method and system based on a dual-tense knowledge graph, and belongs to the technical field of artificial intelligence and knowledge graphs. The method comprises the following steps: through dual-tense decoupling modeling of event effective time and system input time, constructing a versioned knowledge graph and establishing a tense index; efficient retrieval is carried out by adopting a time sequence pruning algorithm based on indexes, and dynamic expansion of a graph pattern is realized through a declarative compiler; realizing abstract-free high-precision result sorting by utilizing a multi-modal concurrent retrieval and attention fusion mechanism; incremental updating of the knowledge graph is realized through streaming data processing and a distributed engine; and closed-loop optimization is formed according to user feedback. The system effectively solves the problems that a traditional RAG scheme is high in data updating delay, tense information is lost, retrieval efficiency is low and expansibility is poor, tense query accuracy, retrieval efficiency and system throughput are remarkably improved, and meanwhile operation and maintenance complexity is reduced.

The invention discloses an equipment network offline event statistical analysis and batch-level stability evaluation method, relates to the technical field of equipment monitoring, and realizes multi-source data time alignment and credibility fusion by constructing a multi-source observation set and generating an equipment context vector and a reference time index. The CUSUM and the hysteretic threshold are combined to identify the offline event, and the real offline and false alarms are accurately distinguished in combination with the characteristics such as the probability type offline duration of the event, so that the event identification precision is improved. Through common cause event cluster clustering and batch multi-dimensional feature extraction, batch-level exposure normalization evaluation is realized. The stability grade and the index contribution degree can be accurately output based on the stability score of the robust normalization and trend degradation item, data support is provided for operation and maintenance decision, and the equipment network operation and maintenance efficiency and reliability are improved. The problems of heterogeneous multi-source data, inconsistent timestamps and missing conflicts in a large-scale equipment network are effectively solved, and the offline event recognition accuracy and the batch-level stability evaluation reliability are improved.

The invention discloses a crane operation controllability health state evaluation method and device in the technical field of engineering machinery, and aims to solve the problem that in the prior art, judgment is performed only by depending on a single type of signals, and collaborative analysis of multi-dimensional indexes is lacked. The method comprises the following steps: in response to a collected sensor signal, carrying out data processing to obtain sensor data; calculating a real-time index according to the sensor data; acquiring a real-time working condition, and dynamically adjusting a scoring weight according to the real-time working condition; calculating a health score according to the real-time index and the score weight; according to the method, various real-time indexes are acquired by adopting various sensor signals, and meanwhile, the score weight is adjusted according to the real-time working condition to acquire the health score, so that the health state evaluation work is cooperatively completed through different indexes, and the method has the adaptive adjustment capability for different working conditions, and is suitable for popularization and application. And the health score in a complex scene is prevented from deviating from an actual state.

The invention provides an ozone pollution prediction method and apparatus, and an electronic device. The method comprises the steps of obtaining initial prediction preparation data of a target area under each time index in at least one time index; determining target prediction preparation data under each time index from the initial prediction preparation data of the target area under each time index; constructing a pollution prediction regression analysis model under each time index based on the target prediction preparation data under each time index; based on the pollution prediction regression analysis model under each time index, a target reference data set is determined, the target reference data set comprises target reference data under each time index, and the target reference data set supports an ozone pollution prediction result used for predicting the target area within the target prediction time. According to the embodiment of the invention, ozone pollution prediction can be conveniently carried out, and prediction cost and prediction time consumption can be reduced.

The invention provides a dynamic extensible time wheel implementation method, assembly, device and medium, and belongs to the technical field of timed task management.The method comprises the steps that a target object is packaged into a time wheel processing object, and the time wheel processing object comprises an object identifier, a polling interval and an optional processing event; storing an object index and a polling time index of a time wheel processing object through a double-hash table structure, constructing a time wheel slot position through an annular time wheel queue, and filling an object identifier of the time wheel processing object into the time wheel slot position; after the time wheel is started, the annular time wheel queue is scanned according to the polling interval of the time wheel, and when the object identifier exists in the time wheel slot position, task operation is executed for the corresponding time wheel processing object; and in the running process of the time wheel, in response to the change request of the time wheel processing object, executing differentiated time wheel updating. Dynamic expansion of a large number of timed tasks is achieved, and the management efficiency is improved.

The invention discloses a data processing method, electronic equipment and a storage medium, the electronic equipment responds to a query request of an index, and determines a target cache region based on timestamp information of the query request of the index; wherein the target cache region comprises at least one of a local cache region and a server cache region; querying in the index data cached in the target cache region to obtain target data; wherein the index data cached in the target cache region is obtained by monitoring operation data sent to the message queue by at least one operation device, and performing real-time index calculation based on index configuration information and the operation data corresponding to the at least one operation device; the index configuration information is used for indicating an index acquisition mode; based on the scheme, the efficiency and the stability of processing the index data can be effectively improved.

The invention provides an alarm processing method based on a rule engine, an enhanced RAG and multi-source data fusion, which comprises the following steps: after responding to a monitoring alarm, extracting a grade, an IP and a trigger text, and carrying out three-level rule verification: if all passes, carrying out automatic self-healing and notifying a person in charge; if not, selecting an optimal case from the RAG library by using a vector similarity + multi-dimensional weighting algorithm, synchronously pulling real-time indexes in a preset time window of the host, fusing the optimal case and the real-time indexes, and inputting the two into a large model to generate a disposal suggestion; if it is suggested that no high-similarity case exists, a multi-role session group collaborative generation scheme is established and stored, the disposal result is used for dynamically updating the knowledge base, the plan is fused when the failure reaches the threshold value, and the successful scheme is written back to the RAG base. According to the invention, efficient automation and intelligentization of alarm processing can be realized, the operation and maintenance efficiency, accuracy and knowledge management capability are improved, and the defects in a complex alarm scene in the prior art are overcome.

The invention discloses an index data analysis method and device, equipment and a program product, and relates to the technical field of data processing, and the method comprises the steps: dividing index data into different hierarchies based on a preset layering strategy, and obtaining a hierarchy label of the index data; obtaining business scene characteristics of the index data, performing scene classification on the index data in combination with a preset index coding rule, and performing physical isolation on a storage result of the index data in a distributed column database physical table of the scene classification based on the hierarchical label; and selecting target index data and data display parameters based on a user interface, and querying the target index data in the physically isolated storage result to generate a real-time index billboard. Through combination of layered architecture design, a scene-based storage strategy and dynamic interaction capability, data islands are eliminated, timeliness is improved, analysis capability is flexible, the requirement of a traditional BI tool for manually configuring a data set is cancelled, and query efficiency is remarkably improved.

Devices and methods for media alignment are disclosed herein. An example device for media alignment includes a media processing device defining a media process path, a drive train, a sensor and a processing platform. The device is configured to receive a waveform signal, identify a first time index of the waveform signal, identify a second time index of the waveform signal, determine a region midpoint between the first time index and the second time index, correlate the region midpoint to a predetermined location index on a web of media, and control the drive train to advance the web of media along the media process path.

The embodiment of the invention discloses a file processing method and device, electronic equipment and a readable storage medium. According to the embodiment of the invention, meta-information, including a time index of a predetermined file and a file type of the predetermined file, of the predetermined file is obtained, a to-be-cleaned file is determined according to a life cycle in the time index of the predetermined file, and then the to-be-cleaned file is cleaned according to the file type of the to-be-cleaned file. Therefore, in the embodiment of the invention, the time index is set for the file with the life cycle, so that whether the file is the to-be-cleaned file or not can be quickly determined only by acquiring the time index of the file during file cleaning, the to-be-cleaned file is convenient to clean, the file cleaning efficiency is effectively improved, and the consumption of computing resources is reduced.

The invention discloses an economic early warning identification method based on big data analysis. The economic monitoring precision and multi-dimensional control are remarkably improved. In the financial dimension, the accumulated income and expenditure difference value and the cost monitoring value are accurately compared, the monitoring threshold value is flexibly calibrated by means of the adaptive cost index, and the economic dilemma is sensitively informed. In the age level, the average age of the group is considered, different age indexes are matched according to age monitoring values to optimize a threshold value, and the economic pressure bearing difference of each age group is considered. And time dimension: locking one-quarter and four-quarter special time periods, finely regulating and controlling by virtue of corresponding time indexes, and conforming to seasonal income fluctuation characteristics. According to the technical scheme, the method comprises the following steps of: selecting a group number, calling a proper personnel index according to the group number, accurately measuring the family scale influence, performing multi-dimensional collaborative operation, breaking through the previous single-dimensional limitation, practically enhancing the accuracy and reliability of economic early warning monitoring, starting from big data and multiple dimensions, and bringing remarkable advantages for an economic early warning monitoring model.

The application discloses a real-time index updating method and an updating device thereof, belongs to the technical field of real-time calculation, can effectively solve the data jitter problem in the real-time index updating process, can effectively solve the data delay problem in the real-time index updating process, is completely transparent to the end user in the process of real-time index updating, and realizes decoupling of the index updating and index use processes; a metadata storage unit is used for maintaining a registration information table of an index set, the registration information table comprises the following fields: a master unit is used for receiving an index set updating request, registering the index set to the metadata storage unit, and submitting a report data calculation job to a Flink cluster.

The invention discloses a time series data analysis method and device, a storage medium and electronic equipment, and relates to the technical field of the Internet of Things. By establishing a composite index structure combining a time index and an equipment index, under the composite index structure, corresponding index entries are dynamically created for a target data flow, the framework of the composite index structure is not changed, and the time series data analysis efficiency is improved. The content is dynamically filled along with the data, and finally a complete composite index result capable of being efficiently retrieved is formed. And the time index adopts a B-Tree structure, so that data in a specific time range can be quickly positioned. Equipment indexing is realized based on a hash table, data generated by specific equipment can be efficiently retrieved, the throughput bottleneck of mass data storage is broken through, and second-level writing and aggregation query of ten millions of data are realized. And in combination with artificial intelligence AI, distributed time series data management and an Internet of Things multi-source data fusion technology, in the face of massive time series data, the analysis processing capability of the time series data is improved to meet the demand of time series data analysis.