296 results about "Data adaptive" patented technology

The invention relates to the field of power data management, and discloses a power big data adaptive management method and system fused with spatial-temporal feature mapping, and the method comprises the steps: obtaining a real-time operation data flow from a multi-source power terminal, and constructing an original power data set with a time sequence label and a device identifier; the method comprises the following steps: dividing an original power data set into parallel processing units based on a storage-while-computing architecture, performing dynamic index updating by adopting an event-driven index mapping rule, and constructing a multi-dimensional data index cache system with real-time responsiveness; identifying a key abnormal trajectory through a time-varying feature nesting mechanism, and performing hierarchical measurement and entropy disturbance analysis on a data fluctuation degree in the key abnormal trajectory by using a streaming feature aggregation network; identifying potential security risk nodes in combination with the structure matching degree between the historical abnormal event evolution graph and the key abnormal trajectory; and generating a multi-level response instruction chain based on the risk assessment result. The method has the advantage of improving the operation safety of the power grid.

The invention belongs to the technical field of low-altitude navigation safety monitoring, and particularly relates to a low-altitude navigation multi-source heterogeneous data adaptive fusion method for multiple scenes, which is a low-altitude navigation multi-source heterogeneous data adaptive fusion method for multiple scenes such as urban air traffic, low-altitude logistics and emergency rescue. Efficient integration of multiple types of monitoring data and position output of the trusted aircraft can be achieved, and accurate data support is provided for low-altitude navigation anomaly recognition and risk deduction. The method comprises the following specific steps: constructing a low-altitude navigation scene classification system and fusion demand mapping, collecting and preprocessing multi-source heterogeneous data, quantifying data credibility and resolving conflicts, constructing a hierarchical adaptive fusion framework and outputting a fusion result. According to the method, efficient and accurate fusion of multi-source data in different scenes is realized by constructing a layered scene adaptation framework and a credibility fusion model, and finally, a high-credibility aircraft position is output, so that reliable data support is provided for low-altitude navigation anomaly recognition and risk management and control.

The invention relates to the technical field of industrial automatic monitoring, and provides a mining equipment remote monitoring method and system and a storage medium. The method comprises the following steps: dividing mining equipment into core equipment and general equipment based on equipment operation parameters; performing fault risk assessment on the core equipment by adopting a dynamic threshold value adaptively generated based on historical data, and performing deviation detection on the general equipment by adopting a fixed threshold value; fusing the fault risk assessment result of the core equipment and the deviation detection result of the general equipment to generate global maintenance decision information; optimizing a field maintenance path according to the geographic position and the emergency degree of the maintenance task; and continuously calibrating system parameters through a closed-loop feedback mechanism. According to the invention, through hierarchical monitoring and adaptive optimization, the problems of uneven distribution of monitoring resources, rigid threshold setting and isolated maintenance decision in the prior art are solved, and accurate configuration of mining equipment monitoring resources and remarkable improvement of fault early warning capability are realized.

The invention relates to the technical field of data analysis, can be applied to intelligent driving, financial science and technology, medical health and other business scenes, and discloses a multi-modal knowledge graph interpretable analysis method, device, equipment and medium, and the method comprises the steps: obtaining multi-modal original data, carrying out semantic analysis, and extracting structured feature information; constructing an entity-associated knowledge graph based on the structured feature information; performing consistency verification by using redundant information of the knowledge graph to complete data quality calibration; setting an abnormal exposure degree and user experience consistency dimension, and constructing a multi-scale aggregation model for dynamically adjusting an index weight; and generating a rating result based on the calibrated knowledge graph and the aggregation model, and performing reverse attribution analysis to generate a source mapping report. According to the method, multi-modal analysis, knowledge graph modeling and dynamic index aggregation are fused, data self-adaptive evaluation and explainable result output are achieved, and analysis accuracy and reliability are improved.

The invention discloses a Direct3D deep cutting behavior compatible method based on an atomization shader, which comprises the following steps: establishing a scene combination metadata table in a DXVK compiling stage, constructing a shader template library containing atomization components and variants, and generating an executable file of the DXVK; after the DXVK is started, deep processing API calling of the D3D application is intercepted, parameters are extracted to generate an atomization parameter set, and scene combination metadata are matched or newly added in a scene combination metadata table to obtain a recommendation strategy and shader component combination; a conversion mode is adaptively selected according to scene metadata to calculate a Vulkan depth value, and a depth comparison operation is adapted to generate a Vulkan depth conversion parameter; the DXVK maps the parameters into DRM depth control parameters irrelevant to hardware and transmits the DRM depth control parameters to a kernel, the DRM packages the DRM depth control parameters and converts the DRM depth control parameters into hardware instructions matched with the GPU, the DXVK loads a shader template and binds the parameters, original logic is spliced to generate a Vulkan shader module, a Vulkan command stream containing the hardware instructions is generated and submitted to the GPU after a pipeline is bound, and finally the GPU executes the Vulkan command stream to obtain a depth rendering result conforming to D3D native logic.

The invention relates to the field of offshore platform monitoring and early warning, and discloses a remote video inspection system and method for an offshore unmanned platform, and the method comprises the steps: a multi-modal data fusion network implementation module obtains multi-modal sensing data; an edge computing node of the edge-cloud co-processing implementation module is responsible for receiving and caching multi-modal sensing data; the self-adaptive task scheduling mechanism module is used for monitoring load states of edge nodes and cloud resources in real time; judging that the task execution position is an edge end, a cloud end or a collaborative mode of edge coarse screening and cloud end fine judgment; when the network bandwidth is limited, the transmission of key alarm data and model updating data is guaranteed preferentially; the intelligent analysis module adopts an improved YOLOv8 target detection model and a multi-modal data fusion network to realize accurate identification and dynamic tracking of equipment abnormity and security risks; and edge end lightweight reasoning and cloud deep reinforcement learning are fused to form a closed-loop inspection flow integrating data acquisition, intelligent analysis and decision feedback.

The invention discloses a multi-modal data adaptive denoising and missing reconstruction method and system, and relates to the technical field of point data denoising and reconstruction, and the method comprises the steps: obtaining to-be-processed multi-modal original data and a modal missing mask; performing unsupervised denoising on image data in the multi-modal original data to obtain a denoised image, and further obtaining multi-modal data; inputting the multi-modal data into a double-flow encoder for processing to obtain a multi-modal embedded vector of cross-modal alignment; the method comprises the following steps of: performing mapping and adding position embedding on a modal embedding vector to obtain each modal coding feature, determining a missing modal based on a modal missing mask, inputting an available modal coding feature into a retrieval enhanced expert model based on prototype memory to perform missing reconstruction to obtain a multi-modal joint representation, and mapping the multi-modal joint representation to a task output space through a full connection layer. Through introduction of unsupervised denoising, double-flow coding alignment and modal knowledge expert hybrid reconstruction, robust representation learning and information complementation under the condition that noise and modal missing exist in multi-modal data are realized.

The invention discloses a carrier tracking method suitable for a multi-bit spread spectrum signal, and relates to the technical field of multi-bit spread spectrum, the carrier tracking method comprises the steps of coarse frequency offset correction, multi-bit spread spectrum, average phase estimation and delay phase correction.The carrier tracking method comprises the steps that firstly, a sliding DFT algorithm is adopted to estimate frequency offset, and a coarse frequency offset available correction signal is obtained; secondly, adaptively selecting the number of available windows according to historical data corresponding to the current signal-to-noise ratio; the method comprises the following steps: firstly obtaining a window intermediate point spread spectrum period, then extracting a carrier phase estimation value of the window intermediate point spread spectrum period, finally delaying a correlation value by the spread spectrum period to realize time sequence alignment and eliminate phase deviation, and finally obtaining information bits corresponding to each spread spectrum period by extracting and correcting a real part of the correlation value, matching a local sequence number and mapping. According to the method, the phase estimation precision under the low signal-to-noise ratio can be improved, parameter self-adaption to high and low dynamic scenes of a satellite is realized, the coherent judgment performance is improved, and the hardware complexity is reduced.

The invention discloses an in-situ coring system and method for a soft rock extension roadway bottom plate. The system comprises a stable anchoring base, a drilling driving and propelling module, a stress adjusting module, a low-disturbance coring module, a rock core in-situ packaging and fidelity module and an intelligent integrated control unit. The stable anchoring base is used for fixing the system on a roadway bottom plate; the drilling driving and propelling module provides rotating power and axial feeding force for the drilling tool; the stress adjusting module is used for adjusting the stress state around the rock core; the low-disturbance coring module comprises a composite function coring drill bit and a double-layer coring pipe, and low-disturbance cutting and containing of a rock core are achieved. The rock core in-situ packaging and fidelity module is used for sealing and packaging the rock core in real time; and the intelligent integrated control unit adaptively controls the coring process according to sensor data. According to the invention, low-disturbance, high-precision and in-situ fidelity rock core acquisition of the soft rock roadway bottom plate can be realized, and the problems of large soft rock coring disturbance, difficulty in fidelity, low intelligent degree and the like are effectively solved.

The invention relates to a pressing plate operation behavior analysis and error prevention method and system based on image recognition and machine learning. The method comprises the following steps: acquiring a pressing plate operation area video stream through image acquisition equipment, and segmenting a video into independent operation event segments based on motion detection and trajectory analysis; extracting key point time sequence data of hands of an operator by using a human body posture estimation model, and constructing a dynamic feature sequence fusing a spatial relationship, kinematics and posture semantics; performing multi-level similarity comparison on the dynamic feature sequence and a standard operation template, and judging operation compliance by a machine learning model in combination with a dynamic threshold value; triggering graded early warning and intervention according to a judgment result; and an incremental learning mechanism is adopted, and the template and the threshold value are adaptively optimized based on historical data. According to the invention, accurate and intelligent analysis and active error prevention of the whole operation process of the pressing plate are realized, and the safety level of electric power operation is effectively improved.

The invention discloses a multipath relay transmission method and system for underground confined space construction data, and belongs to the technical field of wireless communication network routing, and the method comprises the steps: obtaining a signal quality parameter of a communication link and a distribution density parameter of an ad hoc network node, and combining the load state of the node and the information of a neighbor node, decision calculation is executed on a plurality of relay nodes in parallel, relay decision vectors corresponding to the relay decision vectors are generated, a plurality of candidate transmission paths are formed through evaluation, construction data blocks to be transmitted are fragmented and distributed to the corresponding optimal candidate transmission paths to be sent in parallel, all received data fragmented sub-blocks are verified and recombined, and the construction data blocks to be transmitted are obtained. And recovering a complete construction data block. A distributed relay decision based on network state feature fusion, data adaptive fragmentation based on path stability and a multipath parallel transmission and feedback optimization mechanism are adopted, so that the data transmission reliability in an underground complex environment can be improved, the transmission delay is reduced, and the dynamic optimization of network performance is realized.

The invention discloses a student psychological risk perception method based on multiple modes, and relates to the technical field of emotion calculation and intelligent education. The method comprises the following steps: firstly, extracting a facial expression feature vector and a voice intonation feature vector respectively by using a convolutional neural network and Fourier transform through a collected video stream and an audio stream; then adaptive denoising processing is carried out on environmental interference, timestamp alignment and dynamic time warping are carried out on the denoised multi-modal data, time sequence synchronization is ensured, and corrected multi-modal sequence data are formed; then, dynamic emotion track features are extracted from the sequence data, a preliminary emotion state label is generated by comparing the dynamic emotion track features with a baseline threshold value, and the threshold value is adaptively updated in combination with historical data so as to improve the judgment accuracy; and finally, aggregating the emotional state labels of a plurality of students to generate a visual group emotional thermodynamic diagram so as to realize macroscopic perception of group psychological risks. The accuracy, robustness and visualization degree of student psychological state analysis are effectively improved, and an efficient technical means is provided for campus psychological early warning.

The invention relates to the technical field of pile foundation testing, in particular to a test pile load testing method and a loading structure, and the method comprises the following steps: testing preparation: establishing a real-time data acquisition link; graded loading: graded loading: collecting pile top settlement-time sequence data in real time at the load holding stage of the current load stage; self-adaptive load holding judgment: calculating a settlement convergence trend based on settlement data acquired in real time under the current load level; if the prediction model judges that the settlement is fully converged under the current load level, directly entering the next load level without waiting for the settlement rate to be reduced to a fixed threshold value; if the prediction shows that the settlement cannot be converged or tends to be damaged, triggering damage early warning and turning to a damage treatment flow; continuously detecting, namely repeating the step of graded loading and the step of self-adaptive load holding judgment until a preset maximum test load is reached or damage is triggered; and evaluation: evaluating the bearing capacity of the pile foundation. The test efficiency is improved, and the intelligent degree is improved.

According to the special-shaped curved surface laser etching data self-adaptive blocking and track path optimization method, the size of a laser etching block is automatically adjusted according to the curvature radius change of a laser etching pattern, and it is guaranteed that the size of the laser etching data block is smaller than the maximum scanning range of a scanning galvanometer and smaller than the maximum scanning range of the scanning galvanometer; it is guaranteed that the arch height in the blocks is smaller than the focal depth of the etching laser beam, the maximization of the sizes of the blocks is achieved on the premise that the laser etching precision is guaranteed, and the laser efficiency is improved; for a certain laser etching data block, a laser etching track is optimized, so that the idle time of a scanning galvanometer is shortened, and the laser scanning etching efficiency is improved; different laser etching data are partitioned, the moving path of the five-axis moving mechanism is optimized, the idle time of the five-axis moving mechanism is shortened, and the overall efficiency of laser etching is improved.

The invention relates to the technical field of industrial equipment monitoring and defect diagnosis, and particularly discloses a data self-adaptive equipment production process defect identification method and system. By constructing a multi-dimensional delay dynamic reference and based on a joint decision of a delay gradient and a fluctuation entropy, intelligent classification and fast routing of sudden faults, environmental disturbance and composite hidden dangers are realized. According to the system, a three-layer nested diagnosis mechanism of abnormal dimension screening, causal chain tracing and defect mode matching is adopted, a root cause is accurately positioned, and a three-order adaptive response closed loop of instantaneous suppression, parameter compensation and model correction is driven so as to minimize production interruption. Meanwhile, the system periodically calculates the health index of the production line and realizes self-evaluation and collaborative optimization of key parameters, so that a monitoring model can continuously evolve along with the change of the equipment state and the environment, and finally, integration from real-time perception and intelligent diagnosis to adaptive optimization is formed, and the defect identification accuracy, the system stability and the overall operation and maintenance efficiency are remarkably improved.

The invention provides a gradient nano composite high-entropy alloy coating system with a self-monitoring function and a preparation method, and relates to the technical field of material surface engineering.The system comprises a functional gradient coating module used for achieving gradient dispersion distribution of nano ceramic particles on the curved surface of a shield hob so as to form a gradient nano composite high-entropy alloy coating; the fiber bragg grating sensing module is used for collecting strain data of the shield hob through a fiber bragg grating sensor integrated on the shield hob; the surface thermal image monitoring module is used for collecting temperature data and image data of the surface of the shield hob through an infrared thermal imager; and the self-adaptive repairing module is used for determining repairing parameters of the shield hob and generating a repairing instruction, so that self-adaptive repairing is carried out on the shield hob through preset repairing equipment according to the repairing instruction. According to the invention, the sensor network is successfully integrated in the gradient coating, and integrated fusion of structural health monitoring and protection functions is realized.

The invention discloses a multi-source heterogeneous meteorological data adaptive fusion photovoltaic power prediction method, and belongs to the technical field of photovoltaic power generation prediction and meteorological data processing. The method comprises the following steps: acquiring regional meteorological data, site micro-meteorological data and wind cloud satellite secondary cloud picture data, and carrying out preprocessing and feature screening on the regional meteorological data, the site micro-meteorological data and the wind cloud satellite secondary cloud picture data; defining a coupling coefficient vector; calculating fusion temperature, wind speed and irradiance based on the coupling coefficient, and constructing an input feature vector; performing preliminary power prediction by using a time sequence prediction photovoltaic short-term power model and the input feature vector; optimizing a coupling coefficient by adopting a genetic algorithm; and performing final power prediction based on the global optimal coupling coefficient and the time sequence prediction photovoltaic short-term power model. According to the invention, through an adaptive weight optimization mechanism, the problem of insufficient adaptability of fixed weight fusion under different meteorological conditions is effectively overcome, and the engineering landing performance is high.

The invention discloses a code nail stamping defect detection method and system based on machine vision, and belongs to the technical field of industrial automation quality control. The method comprises the following steps: acquiring multi-modal visual data such as a two-dimensional bright field image, a two-dimensional dark field image and three-dimensional contour data of a to-be-detected code nail; preprocessing the multi-modal visual data to obtain bright field, dark field and three-dimensional image data adaptive to a neural network model; inputting the three kinds of image data into a preset neural network model for defect identification processing, and outputting a defect segmentation mask and defect category information; and finally, performing quantitative analysis on the defects based on the output defect information, and judging whether the product is qualified or not according to an engineering specification threshold value. According to the method, multi-dimensional and complementary visual information is fused, and a specially designed neural network model is adopted for analysis, so that the problem of insufficient detection capability of tiny, low-contrast and three-dimensional geometric defects in the prior art can be effectively solved.

The invention relates to the technical field of multi-source heterogeneous data analysis, in particular to a multi-source heterogeneous data-oriented adaptive analysis model construction method and system, and the method comprises the steps: collecting a multi-source heterogeneous data family of a plurality of bolts, and screening alternative diagnosis data to generate a multi-modal key feature vector; judging whether the multi-modal key feature vector is abnormal or not based on the multi-modal key feature vector and the clustering feature vector; obtaining an abnormal multi-modal key feature vector, and generating a fusion feature vector based on a deep learning model; generating an initial diagnosis result based on the fused feature vector; comparing the diagnosis accuracy with a preset threshold value, and judging whether to trigger deep learning model retraining or not; and correcting the attention center of gravity based on the error feature vector and the clustering feature vector, and generating a confidence diagnosis result to position abnormal dimensions and abnormal data. According to the method, the multi-source heterogeneous data analysis efficiency and the anomaly detection accuracy in the bolt tightening process of the battery pack production line are improved.

The invention relates to the technical field of electric power measurement, and discloses an accurate calibration system and method based on electric energy measurement. The method comprises the following steps: preprocessing an acquired original electric signal to generate standard electric signal data; a calibration state identifier is generated through dynamic evaluation, and when fine calibration is needed, data is adaptively segmented into electric signal segments and independent verification and filtering are carried out. Extracting electrical parameter values of the filtered segments, dividing the electrical parameter values into normal and abnormal parameter clusters according to positive and negative deviations of the electrical parameter values and a reference value, and pairing the normal and abnormal parameter clusters into a to-be-verified parameter group after local sorting. And analyzing the relevance of each group by using a pre-established parameter correlation model, counting the number of effective correlation groups, calculating an overall calibration index by combining unpaired parameter values, and judging whether to start calibration or not according to the overall calibration index. According to the method, the signal time period needing to be calibrated can be accurately positioned, invalid calculation is reduced, and the accuracy and robustness of calibration decision are improved through parameter correlation analysis.

The invention discloses a historical block courtyard boundary intelligent extraction and semantic analysis method based on space-air-ground multi-dimensional sensing fusion, and the method comprises the steps: data fusion, the analysis of existing data, and the fusion of multivariate heterogeneous data, and the multivariate heterogeneous data comprises satellite remote sensing, unmanned plane oblique photography and ground mobile laser point cloud. Constructing a cross-scale courtyard enclosure data cube through the fused data, cutting the data cubes of different scales into a plurality of regions, constructing a three-dimensional space cube through the cut regions, and collecting cultural feature information related to historical courtyards through the three-dimensional space cube; according to the method, satellite remote sensing, unmanned aerial vehicle oblique photography and ground mobile laser point cloud are fused through a multi-source heterogeneous data adaptive fusion mechanism, a cross-scale courtyard enclosure data cube is constructed, and a data weighted registration algorithm based on feature significance is proposed. The problems of data missing and registration deviation caused by high building density and serious shielding in historical blocks are solved.

The invention relates to the field of coal mine safety engineering, and provides a coal mine gas control extraction effect dynamic evaluation and closed-loop control method, which comprises the following steps: S1, based on a basic threshold value in an extraction design scheme, periodically and dynamically calibrating the threshold value through a multiple linear regression model, and generating a dynamic threshold value matrix of a core index; s2, collecting multi-dimensional data to construct a fusion data matrix; s3, a dynamic multi-scale convolutional neural network model is constructed and trained, and the model comprises a variable-length filter generator which is used for adaptively generating a variable-length filter according to the input data so as to extract multi-scale time sequence features; outputting a standard state judgment result of the current extraction effect and a core index prediction value of a future preset time period in parallel; and S4, starting a three-level linkage feedback management and control flow coordinated by the AI model and the expert rule base according to the standard reaching state judgment result and the predicted value output in the step S3. Accurate evaluation and short-term prediction of core indexes such as the extraction flow, the gas concentration and the attenuation coefficient are achieved.

The invention discloses a multi-energy management control method for a hybrid power system, and the method comprises the steps: S1, multi-source information processing, S2, working condition prediction modeling, S3, energy distribution optimization, S4, control execution and state monitoring, and S5, model and parameter adaptive adjustment. Future working conditions and required power are accurately predicted through multi-source information fusion and an LSTM model, and a prospective basis is provided for energy distribution. A dynamic weight multi-objective optimization algorithm is adopted, fuel consumption, battery SOC and emission indexes are planned as a whole, an optimal power distribution scheme is solved, and the energy efficiency and the environmental protection property are remarkably improved. Meanwhile, the temperature change rate of the component is calculated in real time through a heat flow evolution algorithm, a cooling system is dynamically adjusted, and precise heat management is achieved. The system further introduces a multi-dimensional deviation closed-loop correction mechanism, key parameters are adaptively adjusted through feedback data, and efficient, stable and reliable operation under different working conditions and environments is ensured.

The invention discloses an early pancreatic cancer prediction and risk stratification system based on artificial intelligence, and belongs to the technical field of medical health data analysis and artificial intelligence. The system comprises a multi-omics data adaptive fusion module, a longitudinal health trajectory coding module, a biomarker combination discovery module, a risk prediction and dynamic layering module and a closed-loop feedback optimization module, and a data confidence index generated by the multi-omics fusion module directly affects the attention weight of longitudinal trajectory coding. Longitudinal track coding adopts a bidirectional long-short-term memory network to extract time sequence characteristics, a biomarker discovery module recognizes a synergistic marker combination through a Transform mechanism, a closed-loop feedback module dynamically adjusts parameters of each module according to a prediction result, and clinical verification shows that the prediction accuracy of the system reaches 85%, the I-stage diagnosis rate is improved by 60%, diseases are discovered 8-12 months in advance, and the diagnosis efficiency is improved. The method is obviously superior to the prior art.

The invention belongs to the technical field of signal detection, and discloses an arrangement optimization method of detection coils for transformer turn-to-turn short circuit fault detection, which comprises the step of uniformly arranging five detection coils in a high-low voltage winding gap corresponding to each coil section along the axial direction so as to ensure efficient detection and accurate positioning of a transformer turn-to-turn short circuit fault. The method is not a pure data black box, but intelligently fuses a physical mechanism and data driving, and ensures that an optimization scheme not only conforms to an electromagnetic physical rule, but also has data self-adaptive capability. The PNN supports end-to-end learning, features can be directly extracted from original signals, the defects of traditional feature engineering are overcome, and the determination efficiency and objectivity of an optimization scheme are greatly improved. Layout optimization and accurate positioning are realized synchronously: in a primary optimization process, the algorithm can accurately position a fault wire turn while outputting an optimal sensor layout scheme.

The invention discloses a high polymer material performance prediction method and system based on deep learning, and the method comprises the steps: constructing an automatic encoder, and carrying out the dimension reduction of various data of a high polymer material through unsupervised learning; optimizing a multi-mode encoder structure by adopting an automatic design mechanism; extracting a fusion characteristic value of the high polymer material by using a multi-modal data encoder; dynamic attention fusion: introducing a dynamic gating weight to adaptively distribute modal weights to input data; introducing physical constraint, and embedding molecular dynamics into back propagation; performing quantum circuit acceleration graph convolution; and predicting and outputting, mapping the fusion characteristic value to the tensile strength and elastic modulus performance indexes of the high polymer material, and realizing nonlinear regression through a multi-layer perceptron. According to the method, a material molecular dynamics equation is converted into a forward propagation kernel from a posterior constraint, the dynamic behaviors of molecules can be simulated and predicted more accurately, and the calculation efficiency is improved while the precision is ensured.

The invention belongs to the technical field of enterprise data management, and particularly relates to an enterprise data visual interaction management system fusing digital twinning and a BI board. Comprising an enterprise multi-source heterogeneous data adaptation access module, a data preprocessing output module, an enterprise entity twin dynamic construction module, a BI visual dimension data aggregation module, an enterprise operation anomaly multi-feature fusion recognition module and a fusion visual interaction presentation module. Influences of data chaos and deviation on analysis results are avoided from the source through an enterprise multi-source heterogeneous data adaptation access module, real-time synchronization of enterprise operation states and early warning of equipment faults are realized based on dynamic digital twinners, and a BI visual dimension data aggregation module meets requirements of users of different levels and reduces docking cost. And by distinguishing positive and negative features and calculating a directional anomaly deviation rate and a comprehensive anomaly degree, misjudgment caused by traditional single features or non-distinguishing directions is avoided, and enterprises are assisted to respond to risks in time and carry out targeted rectification.

The invention discloses a task-driven three-dimensional reconstruction air-ground multi-source data adaptive collaborative acquisition planning method and system, and belongs to the field of three-dimensional modeling. The method solves the problem that an existing three-dimensional model reconstruction method is difficult to consider the modeling coverage range and the precision of local details at the same time when facing a complex environment. According to the technical scheme, the method comprises the following steps of obtaining and preprocessing initial reference data of a target area, performing semantic segmentation and structural analysis on the preprocessed initial reference data, establishing a quantitative acquisition task model based on semantic partitioning and structural analysis results, planning an air-ground collaborative acquisition path for an unmanned aerial vehicle and ground equipment, and establishing an air-ground collaborative acquisition task model for the unmanned aerial vehicle and the ground equipment. The method comprises the following steps: acquiring a current acquired area, estimating the coverage integrity of the current acquired area in real time, completing online quality evaluation, and when an evaluation result shows that coverage vulnerabilities exist, triggering a dynamic re-planning mechanism to carry out supplementary acquisition until the coverage integrity of the current acquired area reaches a preset coverage integrity parameter value in an acquisition task model; the method is applied to three-dimensional modeling.

The invention relates to the field of geographic information system data processing, in particular to a vector tile cutting and publishing system which comprises a heterogeneous data self-adaptive access and normalization subsystem used for receiving multi-mode heterogeneous input; the spatial-temporal feature analysis and index construction middleware is used for generating a metadata index reflecting data spatial distribution features; the vector tile streaming production engine drives a tile generation process based on the metadata index, converts the intermediate state data into a vector tile data packet by adopting a parallel computing architecture, and writes the vector tile data packet into a tile storage warehouse; and the standardized service publishing bus is used for monitoring the change state of the tile storage warehouse in real time and externally providing a vector tile network service interface which accords with a standard protocol. According to the method, the structure barriers of heterogeneous data sources are eliminated, the stability of data production is ensured through topology self-healing, the defect that one set of rules cut all is overcome through a density-based parameter inversion mechanism, and the tile size and the visualization precision are balanced.

The invention discloses a residual service life estimation method and system based on uncertainty calibration, belongs to the technical field of industrial fault prediction and deep learning, and aims to solve the problems of strong spatial correlation, complex time evolution characteristics and lack of reliable uncertainty quantization of prediction results in multi-sensor monitoring data. A residual life prediction framework integrating graph learning, time sequence feature modeling and uncertainty calibration is provided, and the method comprises the following steps: firstly, adaptively constructing a graph structure based on sensor state data, and mining spatial correlation characteristics among multiple sensors through a graph convolutional network; and then modeling is carried out on the time-dependent characteristics in combination with a convolutional long-short-term memory network, and a residual life prediction value and a corresponding variance thereof are output at the same time by adopting a joint probability modeling mode. According to the method, the prediction precision is ensured, the uncertainty quantification with theoretical guarantee is realized, and the credibility and practicability of the residual life prediction result of the industrial equipment can be effectively improved.