40 results about "Time horizon" patented technology

The invention discloses a dynamic environment and multi-event collaborative awareness method based on a multi-modal neural network. According to the method, multi-modal data is obtained through a multi-source sensing device, and a multi-modal data sequence under a unified reference coordinate is formed through time synchronization and space registration. And inputting the data sequence into a multi-modal deep neural network, and adaptively correcting each modal feature weight according to a confidence coefficient change rate between time slices to obtain corrected feature information. And performing external target identification and internal event identification on the feature information to generate a candidate event set, establishing an event priority relationship, and outputting a collaborative event result including an event tag, a spatial position and a time range. And finally, the risk level is determined according to a collaborative event result, and a corresponding early warning or control instruction is generated, so that intelligent identification and collaborative decision-making of multiple events in a complex dynamic environment are realized, and the real-time performance and accuracy of the system in multi-source data fusion and safety monitoring are improved.

The invention discloses an intranet penetration traceability method for multi-scale heterogeneous log behavior analysis, and relates to the technical field of network security, and the method comprises the steps: carrying out the standardization processing of an original log stream of a heterogeneous log source, and generating a unified semantic event stream; performing multi-scale time window division on the unified semantic event stream, calculating an event density characteristic index of each window, and identifying a short-time dense subsequence based on the event density characteristic indexes; clustering analysis is carried out on the identified short-time dense subsequences based on semantic weighted entropy, and aggressive dense segments and normal high-load segments are distinguished; based on the semantic relevance of the event, extracting long-time sparse clues from the unified semantic event stream; a long-time sparse clue is used as a time anchor point, an aggressive dense section adjacent to the time anchor point is positioned in an extended time range of the long-time sparse clue, and an intermediate interpolation event sequence is generated through a conditional time sequence model to reconstruct a complete attack chain.

The invention relates to the technical field of simulation modeling, in particular to a BIM-based building embedded pipeline simulation system, which comprises a pipeline path optimization module, a space conflict detection module, a construction sequence planning module, a dynamic adjustment generation module and a resource utilization rate evaluation module. According to the method, by automatically extracting the spatial distribution information of the pipelines and analyzing the geometrical characteristics and the connection relation of the pipelines, accurate marking of potential conflict positions can be achieved, the comprehensiveness and the accuracy of spatial conflict recognition are effectively improved, and further based on extraction of the construction sequence relation and logic dependency analysis, the construction efficiency is improved. The method can achieve the intelligent judgment of the construction conflict state, achieves the dynamic adjustment of a layout path and a time range, assists in forming a targeted wiring strategy in a conflict region, optimizes the resource space matching degree, automatically recognizes a low-efficiency time period, carries out the real-time identification, improves the response capability of a construction process to the resource utilization efficiency, and improves the construction efficiency. And the precision of path adjustment and the coordination of construction organizations are improved.

The invention belongs to the technical field of water supply network leakage monitoring, and particularly relates to a water supply network leakage node and time window combined positioning method based on a hybrid neural network, which realizes spatial positioning of a water supply network leakage node and time positioning of a leakage time window. Comprising the following steps of 1, collecting node pressure and flow time sequence data of the water supply network in different scenes, and recording leakage nodes and leakage starting and ending time labels; 2, carrying out normalization preprocessing on the data, and constructing a standardized time sequence data set containing a multi-leakage scene mark; step 3, a dual-branch hybrid neural network model TCN-Transform-KAN is built, and a dual-branch hybrid neural network model TCN-Transform-KAN is built; 4, constructing a joint loss function of node classification loss and time regression loss; and step 5, training through a joint loss function constraint model, and outputting a leakage node and a center position and a continuous width of a leakage time window. According to the method, joint precise identification of the leakage nodes and the time range is realized, and technical support of time-space two-dimensional information is provided for intelligent operation and maintenance of a water supply pipe network.

A method performed by a network node for an adaptive prediction of a time horizon for at least one key performance indicator, KPI, in a wireless network is provided. The method includes transmitting towards a user equipment, UE, a machine learning, ML, model to predict the at least one KPI; and receiving from the UE a first message including the predicted at least one KPI and a predicted time horizon for the predicted at least one KPI. A method performed by a UE is also provided.

The invention discloses an individual care product process optimization method and system based on user behavior data, and relates to the field of intelligent data processing, and the method comprises the steps: collecting a historical user behavior data set in a historical time range after a target product is put into the market through multiple channels, and carrying out the data preprocessing and product improvement correlation feature extraction, obtaining a historical product improvement associated feature set; determining an optimal process adjustment time period by combining a predicted product demand growth rate and a predicted product demand fluctuation coefficient within a preset time range; according to the user behavior data set in the optimal process adjustment time period, determining a product improvement index according to the product quality associated feature set; and obtaining an optimal process adjustment scheme by taking approximation to a product improvement index as an optimization target, and carrying out production process iterative optimization within a preset time range. According to the method and the device, the problems of poor user experience, low optimization strategy scene adaptability and inaccurate process optimization direction in the prior art are solved.

The invention belongs to the technical field of computers, and discloses a query method for a closely associated transaction account group, which comprises the following steps of: 1, extracting all connected k-trus communities in a global time interval of an account transaction network, and calculating a real time interval; step 2, carrying out forward evolution refinement; step 3, carrying out backward evolution refinement; step 4, constructing a DAG index structure of the time evolution community; step 5, querying the maximum connected k-trus community of the given account in a specified time interval by using the index; and step 6, returning a query result. According to the method, by constructing the account community structure diagram evolved in the time interval, the repeatedly stored community information is effectively compressed, and the expression ability and query efficiency of account community evolution in a cross-time range are improved. Through recursive stripping and sharing of edge sets among communities, minimum redundant community representation across multiple time windows is realized, and the storage overhead is remarkably reduced while the historical integrity of the communities is ensured.

The invention discloses a hotel reservation management method and device, computer equipment and a storage medium, and relates to the technical field of hotel reservation, and the method comprises the steps: generating and publishing a hotel future option product based on the inventory prediction data of a hotel supplier and a preset price strategy; in response to a purchase request, sent by a user terminal, for future option products of the hotel, receiving earnest money paid by the user, creating an option holding record uniquely associated with the user, and generating an encrypted digital option certificate at the same time; before the option expires or when an agreed performance triggering condition is met, a hotel inventory management interface is automatically called, available hotel room resource information conforming to the check-in time range, the applicable geographic area and the room type level is acquired, and the room type level of the hotel is determined based on a preset multi-factor weighted matching algorithm. And determining a target hotel for the user from the available hotel room source information to complete performance. According to the method, the contradiction between hotel income uncertainty and user travel variability is solved.

A computer-implemented method selects publicly traded assets suitable for trading consideration. A computer accesses market databases, selects assets, and defines an analysis configuration with parameters and timeframes common across assets. For each parameter-timeframe series, the computer forms a normalizing indicator from at least two smoothed sequences and builds multiple deviation bands using specified band families and settings. Within a time window, boundary-crossing events of the indicator are detected for each band. For every asset, the computer records counts of boundary crossings in a result matrix, computes an integral score, optionally with weights, and forms an ordered set of result matrices obtained under alternative configurations. An asset-level output includes a per-matrix score and a set-wise score, such as a sum of per-matrix scores. Assets may be sorted or grouped by these values, and results may be visualized as comparison charts.

The invention relates to the technical field of intelligent traffic scheduling and emergency management, and discloses a multi-rescue center dynamic vehicle path planning method based on double-layer adaptive large neighborhood search. And dividing the global time range into a plurality of continuous time periods, when each time period is finished, freezing the executed path section and the completed task in the previous time period as boundary conditions, and inheriting the real-time position, the residual workload capacity and the unserved accident point set of each rescue vehicle. Through cooperation of a periodic rolling optimization mathematical framework, prediction-driven global reallocation and double-layer ALNS search, a high-quality feasible solution can be continuously output within given time, and time window default is remarkably reduced, commuting distance and total cost are reduced, cross-site resources are more balanced, and online solution is more stable.

This invention discloses a velocity field correction method, apparatus, device, and medium, comprising: determining the initial velocity field of the area to be drilled based on target data of the area; when a new first completed well appears in the area, performing time-depth conversion on the initial velocity field to obtain an initial structural map; correcting the initial structural map based on key point data of the first completed well to obtain a corrected target structural map; determining the velocity correction amount corresponding to each stratum based on the differences between the target structural map and the initial structural map, and the time horizon of each stratum in the target structural map or the initial structural map; and correcting the initial velocity field based on the velocity correction amount corresponding to each stratum to obtain the corrected target velocity field. This invention eliminates well-seismic errors, improves the accuracy of time-depth conversion, and thus obtains high-precision depth-domain seismic data.

Systems and methods for predicting demand in real estate sales receive project-specific input data such as launch pricing, unit mix, project configuration, location factors, and / or market dynamics; identify comparable real estate projects within a defined geographic range based on similarity of project size, configuration, and / or launch conditions; apply an Initial Timeframe Model to forecast the number of units likely to sell within a first timeframe after launch, using a comparative analysis of the identified comparable projects and the project-specific input data; generate a project-specific demand curve for a subsequent timeframe using a demand curve model in which the model establishes a relationship between unit pricing and demand; dynamically refining the demand curve by incorporating real-time market data and / or housing market trends to align predictions with current economic conditions; and provide a graphical representation of the demand curve to guide optimal pricing and sales strategies for sustained revenue generation.

A system for processing and integrating contextual information across multiple time scales simultaneously analyzes immediate and historical data. The system comprises three main components working together to maintain temporal coherence. A temporal processing engine processes input data at different time scales using multiple processors, each dedicated to a specific time horizon. A context integration engine extracts and organizes contextual features from the input data, generating representations for different types of context while maintaining state information across these context types. A fusion engine combines the outputs from the temporal processors and integrates them with the context representations to generate coherent, unified outputs. The three engines operate cooperatively to ensure temporal consistency while processing context across multiple time scales. This architecture enables robust handling of complex temporal and contextual relationships, supporting applications in fields such as customer service, healthcare monitoring, and interactive systems where maintaining coherence across time scales is crucial.

This invention relates to the field of coating production line control technology, and particularly to a time-series linkage control method and system for a coating production line. The method includes the following steps: calculating the congestion level of each workstation; identifying the workstation with the highest congestion level as the bottleneck workstation; calculating the upstream supply segment disturbance index of the bottleneck workstation; obtaining a target time horizon based on the upstream supply segment disturbance index and a preset time horizon; obtaining a buffer target value based on the target time horizon and the real-time processing rate of the bottleneck workstation, wherein the buffer target value is positively correlated with the target time horizon; and updating the buffer capacity in the DBR scheduling algorithm using the buffer target value to control the material input at the beginning of the production line, effectively improving the efficiency of the time-series linkage control of the coating production line.

A system for time-controlled optimization and real-time control of dynamic systems under constraints, based on the time to loss of control, consisting of: a data acquisition module configured to receive real-time data from one or more technical sources representing the state of a monitored dynamic system; a preprocessing module connected to the data acquisition module and configured to process and filter the received real-time data to generate a structured representation of the current system state; a state prediction module connected to the preprocessing module and configured to receive the preprocessed data and, based on the structured representation of the current system state, calculate predicted future states of the monitored dynamic system over a defined time horizon;a Time to Loss of Control (TLC) module connected to the state prediction module and configured to compare predicted future states with predefined thresholds representing acceptable operating limits, determine a Time to Loss of Control (TLC) that represents an estimated time at which the monitored dynamic system is expected to reach an unstable or undesired state, and continuously updates the TLC as new real-time data becomes available;a constraint modeling module configured to identify operational constraints related to the monitored dynamic system, wherein the operational constraints include resource constraints, safety thresholds, and operating rules, and wherein the constraint modeling module is further configured to transform the operational constraints into a structured representation that defines a feasible set of permissible actions;An optimization module connected to the constraint modeling module, comprising a quantum processing unit configured to perform constrained optimization using energy minimization-based computational techniques to determine executable control actions within the remaining time until loss of control, wherein the optimization module is configured to determine one or more executable control actions that improve system stability based on the current system state, predicted future states, determined TLC, and structured representation of the operational constraints, wherein the one or more executable control actions are determined based on the remaining time until loss of control (TLC) such that only actions possible within this time window are considered;A control signal generation module connected to the optimization module is configured to convert the results of the optimization module into executable output signals, which are executed by a control execution module connected to the monitored dynamic system. After the signal is executed, control actions are automatically performed on the monitored dynamic system within the remaining time until loss of control.

The present invention concerns a method for eliminating or mitigating the effect of ionospheric scintillation on GNSS enabled positioning over local, regional and global scale, in areas where static GNSS network providing streaming of S4 and σΦ scintillation indices are available. Depending on the application and on the update rate of the S4 and σΦ streaming by the static GNSS network, they can be used "as is" or can be used as input for the Short-Term Forecasting model (patent application number WO2016185500A1) which forecast the expected S4 and σΦ with a time horizon from seconds up to minutes in advance. Moreover, in the case of very harsh ionospheric conditions which prevent the algorithm from being successful in mitigation approach an alerting is provided to the user.

The embodiment of the invention provides a logistics information dynamic tracking method and device, equipment and a storage medium, and belongs to the technical field of logistics information processing. The method comprises the following steps: obtaining logistics information formed in a fresh food logistics process, and constructing a disturbance fragment used for representing an environmental parameter deviation process; identifying the logistics operation behavior of the fresh food in the time range corresponding to the disturbance fragment, and generating an operation behavior time window aligned with the disturbance fragment; disturbance morphological characteristics used for representing the environmental parameter change process are extracted from the disturbance fragments; and on the basis of the time coverage relationship between the disturbance fragment and the operation behavior time window, and in combination with the disturbance morphological characteristics, determining an abnormal type corresponding to the disturbance fragment, and carrying out association recording on the abnormal type and the corresponding operation behavior information to form a logistics information dynamic tracking result. According to the scheme of the invention, the operation stage positioning and type determination of the abnormal environment in the fresh food logistics process are realized, and the logistics information dynamic tracking result capable of being continuously traced is formed.

The application discloses a kind of thin reservoir top interface accurate structure mapping method and device, belong to oil and gas field development technical field, the method is directed to thin reservoir and innovates top interface seismic horizon correction method, using amplitude variation characteristics eliminates the time horizon deviation error caused by thin layer effect, the amplitude value of formation top interface reflection phase axis and the time deviation of amplitude maximum value change with formation thickness variation, by establishing its function relationship, the relationship between the amplitude value of formation top interface reflection phase axis and the time deviation of amplitude maximum value is directly obtained, to carry out correction.The application avoids the influence of interwell thickness variation on correction effect in the correction process, improves the interpretation horizon precision, and finally improves the precision of thin reservoir top interface structure mapping, and has an important role in finding remaining oil and gas enrichment area in the later development stage.

The invention discloses a converter gas recovery prediction method, device and equipment, a storage medium and a product, and relates to the technical field of ferrous metallurgy, and the method comprises the steps: carrying out the feature extraction of converter gas data collected in real time, inputting the feature data in a first prediction time range into a short-term prediction model, and outputting a short-term prediction value; determining a mid-term prediction value in a second prediction time range according to the production plan and a preset curve template; and performing weighted fusion on the short-term prediction value and the medium-term prediction value according to the overlapping time region, and outputting a total gas recovery amount prediction value and a mixed heat value prediction value in future preset time. According to the method, real-time acquisition and feature extraction are performed on the converter gas data, so that the data are accurately obtained; short-term prediction is carried out depending on high-precision real-time data, medium-term prediction is carried out based on a production plan, and the prediction reliability of a full time scale is ensured through model fusion. And multi-time-scale accurate prediction of gas output is realized.

The application discloses a video retrieval method and system for long and difficult events, and relates to video retrieval. Video frames of a target video are extracted according to different time intervals; a to-be-retrieved event is split according to a preset causal logic split condition; event text is encoded by a text encoder; frame segment sequences are encoded by a video encoder; cosine similarity between a video feature sequence and an event vector sequence is calculated to obtain a similarity score sequence; sliding window analysis is performed on the similarity score sequence, a time range corresponding to a sliding position with the largest similarity score is selected as a target time window; a starting boundary and an ending boundary of the target time window are expanded to obtain a time positioning result of the to-be-retrieved event in the target video; and the positioning accuracy is improved.

The invention discloses a project report generation method and device and electronic equipment. Relates to the field of artificial intelligence. The method comprises: receiving a report generation request of a target project, the report generation request comprising at least one of the following: a time range covered by a to-be-generated project analysis report and a project identifier of the target project; based on the report generation request, obtaining a project management data snapshot meeting the time range, and obtaining target management data; based on the target management data, a model consensus strategy is utilized to generate a project analysis report of the target project, and the model consensus strategy comprises the steps of generating an initial analysis report of the target project based on the N large language models and auditing the initial analysis report. The technical problem that the writing efficiency of the project analysis report is low due to the fact that the writing of the project analysis report depends on artificial experience because the function of the project management tool is solidified in the related technology is solved.

The invention relates to the technical field of budget management, and discloses an intelligent budget monitoring management method and system, which are characterized in that when a budget deviation alarm is monitored, alarm information is not singly displayed, and a basic data unit is accurately positioned through a contribution degree decomposition model, quantitative analysis of a deviation amount, locking of a main cause path and multi-level automatic drilling. Meanwhile, an analysis result is intelligently associated with unstructured information, a key entity and a time range are utilized to retrieve related business events, and financial digits and real business causes are combined to form context insight. Finally, all information is integrated and input into a budget deviation intelligent monitoring engine based on a large language model, a diagnosis report with clear logic and high readability is generated, and budget management and strategic execution capabilities are improved.

The invention relates to a performance curve prediction method and device, relates to the technical field of artificial intelligence, and aims to improve the accuracy of a prediction curve. The method comprises the steps of obtaining first data used for representing a change trend of a historical performance curve of a target object, wherein the change trend of the historical performance curve continuously changes along with time; obtaining second data used for representing attributes of the target object; wherein the attributes are overlapped with the time range corresponding to the historical performance curve, and the attributes influence the change trend of the historical performance curve; encoding the first data according to the second data through the prediction model to obtain a fusion feature, the fusion feature being a feature representation of the historical performance curve in a hidden space after the attributes corresponding to the plurality of historical time periods are fused; and decoding the fusion features through the prediction model to obtain a prediction curve of the target object, the prediction curve being used for predicting the change trend of the performance of the target object.

The invention discloses a fault prediction method and device for financial service, a storage medium and electronic equipment. The method relates to the field of artificial intelligence, and comprises the following steps: obtaining service data of a target financial service in a business platform in N time ranges, the service data at least comprising a service log and service performance data, and N being a positive integer greater than 1; performing feature extraction processing on the service data in each time range to obtain data features of each piece of service data; the data features of the N service data are processed through a fault prediction model, a fault prediction result is obtained, and the fault prediction result represents the probability that the target financial service has different fault types; target information is determined based on the fault prediction result, and the target information represents whether the target financial service breaks down or not and the fault type of the target financial service under the condition that the target financial service breaks down. Through the method and the device, the problem of low accuracy of fault prediction of financial services in related technologies is solved.

The invention discloses an automatic driving decision planning method and related products, and the method comprises the steps: obtaining the target vehicle data of a target vehicle, determining a dynamic time demarcation point according to the target vehicle data based on a preset heuristic function, further generating a short-term trajectory tree according to the target vehicle data, and carrying out the decision planning of the target vehicle based on the short-term trajectory tree. And taking the dynamic time demarcation point as an initial condition to perform expansion processing on the short-term trajectory tree to obtain a long-term trajectory tree, and finally determining a target trajectory of the target vehicle according to the long-term trajectory tree. Thus, the dynamic time demarcation point is determined based on the preset heuristic function according to the target vehicle data instead of a fixed time division mode, the time range of short-term and long-term planning can be adaptively adjusted according to the vehicle state (such as speed) and the surrounding environment (such as traffic density, road curvature and the like), the problem of decision mismatch is avoided, and the accuracy of decision mismatch is improved. And the accuracy and reliability of automatic driving decision planning are improved.

A predictive engine obtains as inputs a set of optimization criteria and a time horizon selection input and applies those inputs to a time horizon model in a variable time horizon prediction engine. A user interface is generated that provides actuatable elements that can be actuated in order to access different time horizon models in the variable time horizon prediction engine. Operational parameters can be adjusted so that the variable time horizon prediction engine provides an output indicative of a control signal that can be used to improve operation of the agricultural system over the selected time horizon.

The embodiment of the application provides a flight guarantee state determination method, device and equipment and readable storage medium, which combines initial processes in the same time period to obtain a combined process, sorts each combined process and the initial process which is not in the same time range of each planned time to obtain a guarantee process of a target flight, and determines the guarantee state of the target flight according to the guarantee process of the target flight and execution data of a processed process of the target flight when the execution data of the processed process of the target flight is acquired. Wherein, the time of each early completion or overtime completion is calculated, and the guarantee state of the flight is determined in real time and accurately on the whole flight guarantee state level.

A computer-implemented method, a computer program product, and a computer system for retraining a model in case of a drift in machine learning. A computer detects a drift in machine learning. A computer identifies in a database features and a response of a machine learning model. A computer determines a time window of the drift. A computer extracts, from the database, data of the features and the response in the time window. A computer determines whether extracted data is sufficient for retraining the machine learning model. A computer, in response to determining that the extracted data is not sufficient for retraining the machine learning model, interpolates one or more of the features for a predetermined future time horizon. A computer interpolates a response corresponding to one or more interpolated features. A computer retrains the machine learning model, using the one or more interpolated features and an interpolated response corresponding thereto.

Technologies are disclosed herein for distributing information. The disclosed technologies determine an application element configured to receive information that is updated at a variable rate, the information pertaining to an object. Feature data is received that is associated with the object and data associated with use of the application element. The feature data includes a time horizon for the object and supplemental information associated with the object. Based on the feature data and the data associated with use of the application element, a first rate is predicted for sending the information about the object to the application element. The information is sent to the application element at the first rate.