387 results about "Dynamic modelling" patented technology

The invention discloses a public opinion event multi-mode semantic fusion modeling and abstract generation method and system, and relates to the field of natural language processing and social network analysis. Through the multi-mode semantic fusion technology, the short text understanding ability is improved, and the problems of semantic fuzziness and network language diversification are solved. Meanwhile, through a cross-window event cluster matching technology, an event evolution path with time continuity is constructed, and comprehensive capture of event dynamic characteristics is realized. Besides, the structured event abstract is automatically generated by utilizing the generative model, so that the consistency and the information density of the abstract are improved, and the actual application requirements are met. Through the innovations, the defects in the aspects of semantic comprehension, dynamic modeling and abstract generation in the prior art can be effectively overcome, a more efficient and accurate solution is provided for monitoring and analysis of public opinion events, and the method has wide application prospects in the fields of public opinion monitoring, emergency early warning, social media data analysis and the like.

The invention relates to the technical field of data center dynamic modeling, and discloses a twin model dynamic modeling method for a data center DCIM platform, which comprises the following steps: constructing a discrete state space model containing a thermal coupling matrix and a system matrix, collecting real-time power and temperature time sequence data, and calculating a cross-correlation function to lock hot air dynamic transmission lag time; calculating cut-off frequency based on physical attributes of the cabinet and decomposing data into high and low frequency components by using a complementary filter; according to the method, the model parameters are made to return to a physical source through a frequency domain decoupling mechanism, the problem of aliasing of airflow coupling and structural thermal inertia parameters in a traditional single-scale identification method is solved, and the method is suitable for large-scale identification. And the physical authenticity and prediction robustness of the twin model under a complex working condition are improved.

The invention relates to the technical field of power system automation, and particularly discloses a cross-domain heterogeneous computing power real-time calling and unified scheduling system for the power industry, which comprises a computing power resource global sensing unit, a business demand dynamic modeling unit, a cross-domain unified scheduling decision unit and a task execution and feedback control unit, according to the method, heterogeneous computing power states are collected in real time, task requirements are dynamically modeled, a scheduling scheme is generated based on multi-objective optimization, and task execution monitoring and load balancing are realized by means of closed-loop feedback, so that the utilization efficiency of global computing power resources is improved, and the real-time performance and reliability of power business are guaranteed.

The invention relates to the technical field of three-dimensional modeling, in particular to a dynamic modeling method for a geological structure three-dimensional model. The method comprises the steps that multi-source data are acquired and preprocessed, a voxel semantic fusion algorithm based on variational optimization is introduced, after semantic information is extracted from the preprocessed multi-source data, the preprocessed multi-source data are mapped to a target three-dimensional space grid, and an optimal semantic fusion vector is obtained; based on the optimal semantic fusion vector, generating a standard voxel data pool, and constructing a geological structure three-dimensional model; monitoring data change, calculating the position of a newly added data point, combining the standard voxel data pool to obtain a space updating area, and modeling the space updating area to realize model updating; and after the modeling of the space updating region is completed, optimizing the boundary continuity. The problems that multi-source geological data cannot be directly used for structure construction and semantic fusion of a three-dimensional model, dynamic response to newly-added data is lacked, and geometric discontinuity and structural logic discontinuity exist at the boundary are solved.

The invention relates to the field of roadway surrounding rock deformation analysis, in particular to a roadway surrounding rock deformation intelligent monitoring method and device based on three-dimensional laser scanning. The method comprises the following steps: performing three-dimensional laser scanning on a roadway to generate equalized laser point cloud data; performing point cloud relative deformation detection and surrounding rock deformation dynamic modeling on the equalized laser point cloud data to construct a dynamic surrounding rock deformation model; performing space grid division on the dynamic surrounding rock deformation model, and performing differential change trend analysis to generate a multi-unit deformation trend; performing nonlinear evolution analysis and deformation situation prediction in a future time period on the multi-unit deformation trend to generate a deformation prediction result; and performing deformation situation interference simulation on the deformation prediction result, and extracting various interference simulation results. According to the invention, surrounding rock deformation data is accurately obtained in real time, and the deformation trend is dynamically analyzed and predicted, so that early warning is performed in advance, the safety risk is effectively reduced, and the safety and monitoring efficiency of underground engineering are improved.

The invention is suitable for the technical field of road construction, and provides a wind power plant road intelligent line selection method based on multi-dimensional constraint dynamic modeling, and the method comprises the steps: obtaining a multi-dimensional constraint condition of wind power plant road line selection, and constructing a constraint model; generating a slope grid map based on a digital elevation model, and fusing the slope grid map with a constraint map layer reflecting geological stability and a land cover type to generate a dynamic cost surface; adopting an improved heuristic search algorithm to perform parallel path search from the road starting point to each target fan position, generating a plurality of initial feasible paths, and calculating a comprehensive cost value; screening a candidate line set according to a comparison result of the comprehensive cost values; receiving real-time updating data in a construction process and environment monitoring, inputting the real-time updating data into the constraint model to trigger model updating, performing iterative optimization on the candidate line set based on the updated constraint model, and outputting a final line selection scheme; and the line selection efficiency, economy and reliability are effectively improved.

The invention relates to the technical field of dynamic modeling and digital twinning, and discloses a digital twinning dynamic modeling method for new material research and development, which comprises the following steps: constructing a low-order linear reference model, calculating a residual signal flow, and extracting residual non-statistical topological characteristics by using a sliding window to retrieve a system matrix correction increment; constructing a parallel shadow verification model to execute virtual deduction, and screening and executing parameter hopping according to a convergence divergence ratio; according to the method, through a closed-loop mechanism of residual form fingerprint identification and shadow deduction verification, the problem of capturing discontinuous physical phase change under low computing power is solved, and real-time high-fidelity synchronization of the model on complex microscopic evolution is realized while environmental noise interference is suppressed.

The invention relates to the technical field of power system safe operation safety and intelligent monitoring, and discloses a vegetation intrusion risk assessment method and system for a power transmission corridor and electronic equipment, and the method comprises the steps: obtaining multi-source remote sensing data and meteorological data of vegetation in a preset range of a target power transmission corridor; performing data fusion on the multi-source remote sensing data to obtain fused data; extracting a vegetation height change sequence based on the fused data, and performing prediction based on the vegetation height change sequence and the meteorological data to obtain a future time sequence height change curve; obtaining engineering parameters of a target power transmission corridor, and calculating a vegetation invasion risk value based on the engineering parameters and a future time sequence height change curve; and performing risk assessment based on the vegetation invasion risk value. The method breaks through the limitation of traditional static monitoring, integrates InSAR time sequence analysis and a machine learning algorithm, realizes vegetation growth dynamic modeling and future trend prediction, realizes risk level quantitative evaluation in combination with power transmission line structure parameters, and can provide technical support for power grid risk early warning.

The invention relates to an intelligent work order automatic creation and circulation system and method based on context awareness, provides dynamic modeling, and relates to the field of systems or methods specially suitable for administrative, commercial, financial, management or supervision purposes. The system comprises a work order analysis device which is used for dynamically modeling a convolutional neural network model, and intelligently analyzing a creation demand and a creation type of a skill work order of each consultation work order according to consultation associated data of each consultation work order in a recent time interval; and the automatic creation device is used for executing automatic creation of a subsequent skill work order based on the intelligent analysis result. In order to solve the technical problem that the skill work order of the tourism e-commerce is difficult to synchronously analyze the creation demand and the creation type, an artificial intelligence model obtained by dynamic modeling of the tourism e-commerce is used, and according to context sensing data of a visitor in each consultation work order, synchronous analysis of the creation demand and the creation type of the subsequent skill work order is realized; therefore, the technical problem is solved.

The invention discloses a heterogeneous network embedding method based on dynamic heterogeneous network decomposition and a hierarchical space-time attention mechanism. The method comprises the following steps: dynamic network decomposition: decomposing a heterogeneous network sequence into independent subgraph sets according to edge types; multi-order node attention aggregation: for each sub-graph, fusing information of nodes and different edge types; semantic level attention is embedded, wherein global semantic embedding is generated in combination with global relation type weight and meta-path long-short-term memory network coding; time dynamic modeling: respectively calculating a time attention weight and a Monte Carlo sampling approximate Horkes intensity, and then aggregating and splicing results of the time attention weight and the Monte Carlo sampling approximate Horkes intensity to obtain a final embedding result. According to the method, through cross-granularity semantic modeling and efficient time sequence dependence learning, an innovative solution is provided for node representation of the dynamic heterogeneous network.

The invention provides a multi-mode fusion laser directional energy deposition real-time monitoring method and system, and belongs to the technical field of additive manufacturing. Comprising the steps that real-time data collection is conducted on a molten pool area in the laser directional energy deposition process through synchronous triggering of a multi-source sensor; a distributed clock network is constructed, environment influence parameters are collected in real time in combination with a working condition sensing unit, dynamic modeling and correction are conducted on asymmetric link delay in the clock synchronization process, and space-time alignment of multi-source sensor data is achieved; after the multi-source sensor data subjected to space-time alignment is preprocessed, data-level fusion, feature-level fusion and decision-level fusion are performed in sequence. The state of the molten pool can be accurately monitored in real time, and the quality and stability of the laser directional energy deposition process are improved.

The invention relates to a knowledge graph construction and retrieval optimization method based on water affair data, and the method comprises the steps: taking the unified knowledge expression of a water affair object and an event thereof as a target, determining a water affair business field and business range, and constructing a water affair knowledge graph conceptual model covering a static knowledge ontology and a dynamic event ontology; performing knowledge extraction on the obtained multi-source heterogeneous water affair data, converting the data into a unified triple form, and constructing a data layer of a water affair knowledge graph; and on the basis of the data layer, node, path and region query is performed on the knowledge graph by adopting a graph algorithm, and semantic enhancement retrieval and reasoning are performed in combination with a large language model. By constructing the water affair knowledge graph and introducing knowledge extraction and retrieval enhancement capability of a large language model, dynamic modeling and semantic reasoning of water affair objects, events and relationships thereof are realized, high-risk area identification, emergency path planning and emergency early warning are supported, and the knowledge utilization rate and intelligent decision level in the water affair field are improved.

The invention discloses a source load impact and system bearing life evaluation method and system, equipment and a medium, and relates to the technical field of traction power supply, and the method comprises the steps: collecting data to recognize impact, building a source load model, analyzing an impact diffusion path, evaluating system bearing, fusing multi-dimensional damage, calculating the residual life of the equipment, and automatically updating parameters. And generating a closed-loop regulation and control strategy. The system comprises a data analysis and dynamic modeling module, a bearing evaluation and conduction analysis module, a damage coupling and life prediction module and a parameter correction and strategy generation module. The traction impact time domain-frequency domain-topology integrated quantization is realized, and the problem of impact time-space diffusion assessment is solved. A life model is constructed, and multi-dimensional stress damage is accurately mapped. The micro-simulation and on-line filtering technology is utilized to ensure that model parameters are adaptively updated along with working conditions, and the problem of disjunction between evaluation and actual conditions is solved. The method supports side cloud cooperation and closed-loop pre-control, effectively improves the safety margin of the power grid, and delays the aging of equipment.

The invention provides a data-driven saline screw compressor unit dynamic modeling method, and belongs to the technical field of industrial intelligence and predictive maintenance. According to the method, adaptive deep denoising of original data is realized through a composite signal processing flow of fusing variational mode decomposition, permutation entropy criterion and wavelet packet transformation optimal threshold denoising; meanwhile, a sectional sampling strategy is introduced to enhance the diversity of training data. Then, a wavelet multi-scale energy entropy extraction layer is used for constructing a high-information-density feature vector; furthermore, a prediction model formed by multiple layers of stacked long and short-term memory network units is adopted, and the complex time sequence dependency relationship of the system is deeply captured. According to the method, pure and stable system dynamic representation can be extracted from high-noise industrial data, the dynamic characteristics of the system in the full working condition range are accurately described, and it is ensured that the prediction result is self-consistent physically and reliable in engineering, so that the prediction precision and generalization performance of the model are remarkably improved.

The invention discloses an ecological restoration dynamic modeling system for a power grid project slash in a high-altitude area of the Tianshan Mountain in Xinjiang, and the system comprises an obtaining module which is used for obtaining historical multi-source heterogeneous ecological data of the power grid project slash in the high-altitude area of the Tianshan Mountain in Xinjiang Mountain, and carrying out the standardization processing of the historical multi-source heterogeneous ecological data to generate an ecological data set; the historical multi-source heterogeneous ecological data comprises climate data, soil data, hydrological data and vegetation data; the model construction module is used for constructing a climate-soil-hydrology-vegetation four-dimensional coupling ecological restoration dynamic prediction model based on the ecological data set; the dynamic simulation module is used for presetting an ecological restoration scene; inputting a preset ecological restoration scene into the four-dimensional coupling ecological restoration dynamic prediction model for dynamic simulation prediction, and when a dynamic simulation prediction result meets requirements, obtaining an ecological restoration dynamic prediction model of the power grid project slash in the high-altitude area of Tianshan Mountain in Xinjiang; and carrying out accurate dynamic simulation and prediction on the ecological system of the project slash, and formulating and adjusting a scientific full-period management and protection scheme.

The invention provides an intelligent construction site safety monitoring and early warning method and system, and the method comprises the steps: fusing the multi-source data, such as personnel positioning, equipment state and environment parameters, and building a structured time sequence feature set with spatial semantic annotation; constructing a symbolized rule base in combination with a field safety specification; applying a dynamic graph attention network, dynamically modeling personnel, equipment, environment nodes and an interaction relationship thereof, and deducing potential causal association strength between spatial entities; through path semantic distillation and anti-fact intervention, a compliant and accurate multi-stage causal reasoning path is formed, and key risk nodes and propagation channels are visualized in a three-dimensional BIM scene; and finally, dynamic early warning event generation and natural language interpretation are realized, model closed-loop optimization based on field feedback is supported, and the timeliness, interpretation and adaptive capability of construction site safety risk discovery are improved.

The invention discloses an intelligent data question and answer method and system based on index semantic enhancement, and relates to the technical field of intelligent data question and answer. The method comprises the following steps: acquiring a data analysis demand in a natural language form; the method comprises the following steps: performing information extraction on a data analysis demand by utilizing an index semantic enhanced large language model, generating a query statement according to the extracted information, recalling and rearranging most relevant indexes and elements in an index semantic knowledge base by utilizing a knowledge retrieval enhancement layer according to the data analysis demand and a historical session on the basis of mixed retrieval, and generating a query statement according to the query statement. Generating a prompt template according to a retrieval result of the index semantic injection layer by utilizing the index semantic injection layer, and performing semantic understanding reasoning according to the prompt template to obtain corresponding information; and checking compliance according to the query statement, and returning a query result. The method aims at achieving efficient mapping, dynamic modeling, intelligent SQL generation and optimization, authority safety management and control and multi-round intelligent interaction of natural languages and enterprise-level index assets.

The invention discloses a multi-parameter shallow gas detection method which comprises the following steps: S1, self-adaptive sensing network layout and multi-parameter synchronous acquisition: monitoring nodes synchronously acquire underground electrical parameters, soil gas composition and concentration, soil mechanics and pore pressure parameters and an environmental noise background value; accessing a hydro-meteorological data stream of the target area in real time; s2, multi-source interference signal identification and data purification: obtaining a purified multi-parameter data sequence reflecting real geological anomalies; s3, water-gas-force coupling dynamic modeling and migration simulation: outputting a shallow gas migration risk thermodynamic diagram; s4, intelligent diagnosis based on transfer learning and causal discovery: performing time sequence analysis on the purified multi-parameter data sequence, and identifying a gas reservoir core area, a dominant migration channel and dominant control factors influencing gas aggregation and migration; and S5, dynamic risk assessment and graded early warning: quantitatively calculating the risk probability.

The invention discloses a strip mine slope weak plane dynamic modeling system based on real-time working condition feedback, and belongs to the technical field of geographic model 3D dynamic modeling of computer graphics. Comprising a data acquisition module, a data preprocessing module, a modeling and updating module, a working condition feedback calculation module and a working condition feedback adjustment module which are connected in sequence, the method comprises the following steps: calculating an overlapping region and an adjacent overlapping region of a working condition operation influence region and a strip mine slope three-dimensional model, and intensity data of coordinate points of a three-dimensional weak surface under the influence of industrial and mining operation, and labeling and dynamically updating to generate inspection information of the overlapping region and the adjacent overlapping region; meanwhile, the three-dimensional model of the strip mine slope with more accurate dangerous area form is reconstructed and updated, the accuracy and timeliness of weak plane stability analysis and early warning are improved, and the problems that in the prior art, serious defects exist in the aspect of associating real-time mining working conditions, weak plane stability analysis lags behind, accuracy is low, and early warning accuracy and timeliness are low are solved.

The invention discloses an end face tool sharpener control system, and relates to the technical field of numerical control systems. The end face tool sharpener control system comprises a data preparation and recording module, a tool type parameter dynamic modeling module, a multi-axis linkage control module, a grinding wheel pressure self-adaption control module, a grinding wheel abrasion compensation module and a technological parameter optimization module. By comprehensively collecting multi-dimensional historical processing data and constructing a set framework, automatic classified storage of parameters is realized; the NURBS curve interpolation algorithm is adopted, and the machining precision and smoothness of the complex shape are guaranteed; instructions are synchronized in real time, errors are compensated, vibration is monitored in combination with an acoustic emission sensor, and machining stability is improved; a fuzzy PID model and machine learning are utilized to dynamically adjust parameters and optimize the parameters in advance; through machine vision and digital twinning, abrasion is accurately recognized, and the finishing period is predicted. And based on the hybrid model and reinforcement learning, predicting a processing result, dynamically adjusting parameters, and performing error compensation and sensitivity analysis.

The invention provides a mining disturbance surrounding rock bearing structure analysis method and a mining disturbance surrounding rock bearing structure analysis system, and relates to the technical field of surrounding rock bearing structure analysis. Through intelligent risk prediction of deep learning methods such as LSTM and the like, whole-process refined management and control of the surrounding rock bearing structure under roadway mining disturbance are realized; in combination with fault zone distribution, rock mass damage evolution and disturbance working conditions, the scheme can identify potential damage surfaces in real time and dynamically correct key geomechanical parameters, and it is ensured that the supporting structure design is highly matched with the actual risk; through hierarchical risk early warning and multi-objective optimization measures, the timeliness and accuracy of disaster early warning are improved, and accurate configuration of support resources and collaborative minimization of economic loss are effectively realized, so that the safety guarantee and management efficiency in a complex underground engineering scene is enhanced.

The invention discloses a collaborative recovery effect evaluation method for transition of campus buildings and landscapes. According to the method, campus environment data, human body physiological response data and psychological perception data which are aligned in time and space are acquired through gradient path design and multi-modal data synchronous acquisition; performing deep analysis on the environment data, extracting visual and auditory features, and constructing a standardized environment feature tensor; based on the environment characteristic tensor and the physiological response data, carrying out restorative effect dynamic modeling, and quantifying a nonlinear mapping relation and a cross-sensory synergistic effect between an environment factor and a restorative effect; and according to the output of the dynamic model, carrying out multi-objective optimization to generate an environment reconstruction scheme, realizing parameterized scene reconstruction and effect verification by using generative artificial intelligence, and constructing a dynamic recovery evaluation map of the campus space. According to the method, scientific quantification and visual evaluation of the restoration potential of the campus transition environment are realized, and decision support is provided for refined environment design.

The invention discloses an intelligent driving end-to-end decision model iteration method and device, equipment and a medium, and relates to the technical field of intelligent driving, and the method comprises the steps: collecting multi-sensor scene state data, carrying out the preprocessing, and generating a high-dimensional feature vector; mapping the high-dimensional feature vector from an original space to a bird's-eye view angle space through a BEV fusion model, and performing feature interaction in the bird's-eye view angle space by using an attention mechanism to generate high-dimensional environment features; performing dynamic modeling on the high-dimensional environment features by using a graph neural network and multi-task learning, and outputting an environment perception graph; and designing a reward function including safety, driving efficiency and navigation matching degree by using a Markov decision process based on the environmental perception map, so as to optimize the constructed decision model based on a PPO algorithm, and continuously adjusting through multiple iterations until the decision model converges. According to the invention, through combination of multi-sensor data fusion and deep learning, efficient and steady intelligent driving decision in a complex environment is realized.

The invention discloses a highway network traffic flow detection and congestion situation dynamic modeling system, and relates to the technical field of highway traffic management, the highway network traffic flow detection and congestion situation dynamic modeling system comprises a sensing layer, a transmission layer, a first processing layer, a second processing layer and an application layer, the sensing layer deploys multi-modal detection equipment to realize traffic flow data acquisition of highway key nodes; the transmission layer is provided with an edge server cluster for realizing real-time transmission of data acquired by the sensing layer; the first processing layer is used for realizing an RGBT multi-mode traffic flow detection subsystem, the second processing layer is used for realizing GNN congestion situation dynamic modeling, and the application layer provides visual display and decision support for a traffic management department. The RGBT detection technology and GNN modeling need to be deeply fused, a perception-modeling-analysis-early warning closed-loop system is constructed, the requirements of a traffic management department for accurate management and control and efficient scheduling are met, and the practical requirements of current highway management pain points are met.

The invention discloses an intelligent surveying and mapping paying-off system based on BIM model and robot cooperation, and relates to the technical field of engineering surveying and mapping and intelligent construction, and the system comprises a multi-source sensing collection module which synchronously collects construction data and fuses and outputs a high-precision result; the BIM dynamic modeling module receives the fusion data to generate a dynamic model and updates the dynamic model; the multi-dimensional comparison and safety map module realizes deviation analysis and early warning and constructs a safety grid; the intelligent decision planning module calculates compensation parameters and generates a decision scheme; the construction execution feedback module works according to the scheme and feeds back data to form a closed loop to adapt to construction requirements; according to the invention, through multi-source sensing and BIM dynamic modeling, data depth integration and model real-time adaptation are realized, and accurate guidance is provided for paying-off operation; a three-dimensional safety grid is constructed to plan an optimal path, and the operation stability and accuracy are improved by matching with a closed-loop adjustment process; the system joints are connected with all links, interference is reduced, and efficiency and quality are improved.

The invention relates to the technical field of coal mining, in particular to a coal mining prediction method and a coal mining prediction device. The method comprises the steps of obtaining geological basic data, and obtaining a regional geological database based on the geological basic data; based on the regional geological database, carrying out three-dimensional geological modeling by adopting a multi-field coupling modeling method; performing dynamic iteration correction on the three-dimensional geologic model based on the data consistency constraint of the virtual sensor node; production working condition parameters of the working face in the mining process are obtained in real time, the production working condition parameters are converted into constraint conditions of the three-dimensional geologic model, and dynamic modeling in the mining process of the working face is carried out; based on the production working condition parameters and the three-dimensional geologic model, intelligent analysis and multi-field coupling simulation are carried out, and a mining prediction result is obtained; updating the mining decision in real time based on the mining prediction result; the device is applied to the method, and accurate prediction, dynamic optimization and safety guarantee of the mining process are achieved.

The invention relates to the technical field of action recognition, in particular to an AI-based motion posture recognition system, which comprises a key point extraction module, a graph structure construction module, a node trajectory modeling module, an action segmentation analysis module and a posture recognition output module. According to the method, the spatial position of each node of the human body in the continuous image under the time sequence structure is extracted, and the graph structure sequence based on the topological relation between the nodes and the motion trend is established, so that the dynamic modeling and the structure expression of the skeleton motion can be realized, and the trajectory characteristics of node position change, speed fluctuation, direction change and the like are analyzed; according to the method, key paragraphs with inversion and abnormity in actions are effectively identified, action segmentation classification is carried out in combination with frame sequence spacing and key node activity states, accurate analysis of complex action processes is realized under the condition of not depending on external hardware, and the accuracy and timeliness of posture identification are improved in a mode of matching with a standard template.

The invention discloses a multi-source fusion deep learning visual navigation method, system and device and a storage medium, and the method comprises the steps: collecting a visual image sequence and a Beidou positioning signal obtained by a robot in an operation process, carrying out the preprocessing, and generating a fusion input sequence with a quality score; performing two-channel feature coding on the visual image sequence and the Beidou positioning signal, mapping the features to a unified joint representation space, and performing dynamic modeling on the reliability of each mode in the current environment; based on the confidence coefficient of each mode and the cross-mode correlation, generating a double-branch dynamic mask of vision and Beidou, and combining a confidence coefficient modulation vector to fuse the two types of features to obtain a fused feature sequence; and performing context modeling on the fused feature sequence, constructing a joint optimization objective function for cooperative training, and outputting a robot navigation trajectory. According to the invention, the navigation precision, reliability and autonomy of the robot in a complex environment can be improved.

The invention belongs to the technical field of intelligent oil field and oil and gas yield prediction, and particularly discloses an oil well yield prediction method and system based on a frequency domain enhanced BiLSTM-Transformer hybrid model, and the method comprises the steps: collecting oil well historical production time sequence data, and carrying out the abnormal value elimination and normalization preprocessing; constructing a multi-scale feature system, and integrating a time domain lag term, a rolling statistic, and frequency domain dominant frequency and amplitude features extracted through Fourier transform; introducing an oil well embedding vector to explicitly compensate the inter-well static difference; the features are input into a BiLSTM-Transform hybrid model, a local dynamic state is captured by the BiLSTM, and a global trend is modeled by the Transform; an oil well level adaptive optimization strategy is adopted, sample imbalance is relieved through sample dynamic weighting and adaptive learning rate scheduling, and the generalization ability is improved in combination with gradient cutting and an early stop mechanism; and finally outputting an oil well yield prediction value. According to the method, the problem of prediction deviation caused by yield sequence non-stationarity, inter-well difference and sample imbalance is effectively solved, and the prediction precision is remarkably improved.

The invention discloses a multi-energy cooperative power generation prediction method and system based on multivariate data fusion, and relates to the technical field of water conservancy and hydropower engineering, and the method comprises the steps: collecting meteorological, hydrological and load data, executing hour-by-hour alignment and abnormity correction, and building a wind-light-water multi-source data set; according to a low ebb, noon and peak division rule, calculating a time-sharing mean value of each energy source and performing seasonal correction to generate a power generation factor; and executing multi-energy collaborative prediction and scheduling based on the power generation factors, and updating power generation parameters by adopting a rolling verification and credibility mechanism. According to the method, the time sequence consistency and reliability of input data are improved, power generation factors are generated through time-sharing mean value calculation and seasonal correction, dynamic modeling and periodic correction of energy output characteristics are achieved, a dynamic prediction and self-adaptive updating mechanism of wind, light and water complementation is established, and the power generation efficiency is improved. Therefore, the accuracy and the scheduling response capability of multi-energy cooperative power generation prediction based on multivariate data fusion are improved.