19 results about "Gradient analysis" patented technology

The invention relates to the technical field of soil pollution investigation and environment monitoring, and discloses a soil pollution low-carbon investigation system, which comprises a data acquisition and preprocessing module for acquiring site multi-source heterogeneous data and performing data quality evaluation and preprocessing; the feature extraction and fusion module is used for performing feature extraction and feature fusion; the pollution prediction and uncertainty quantification module is used for carrying out full-field pollution prediction and uncertainty quantification; the sampling optimization module is used for carrying out sampling value evaluation and gradient analysis; the carbon footprint monitoring module is used for carrying out carbon emission calculation based on an identification result and carrying out real-time monitoring on the carbon footprint; the path planning module is used for planning a sampling path and deploying an edge reasoning system to carry out real-time dynamic path decision; the scheme decision-making module is used for generating investigation candidate schemes and selecting an optimal scheme according to decision-making rules; according to the method, a multi-scale depth feature extraction and cross-modal fusion technology is adopted, and the collaborative value of multi-source heterogeneous data is fully mined.

The application relates to a cooking utensil state detection method, an intelligent range hood and a computer device, wherein the method comprises the following steps: performing infrared thermal imaging on a target region above a heated cooking utensil to obtain a real-time temperature matrix corresponding to the target region; the real-time temperature matrix comprises a plurality of temperature values, and each temperature value corresponds to a corresponding position in the target region; gradient analysis is performed on each temperature value in the real-time temperature matrix to obtain a corresponding gradient amplitude matrix; the gradient amplitude matrix comprises a temperature gradient amplitude corresponding to each temperature value; based on each temperature value in the real-time temperature matrix and the gradient amplitude matrix, a heat diffusion coefficient of the cooking utensil is determined; and based on the heat diffusion coefficient, the current state of the cooking utensil is identified. Through the application, the problem that the actual state of the cooking utensil cannot be accurately detected is solved, and the state of the cooking utensil is accurately detected, so that the user can timely avoid the safety and health risks caused by the change of the state of the cooking utensil.

This invention discloses a digital twin management and inspection method and system for gas risers. The method establishes a riser feature map by performing a weighted analysis of missing data from survey records and geospatial data. It generates a twin archive set of shading depth data through gradient analysis of facade visibility angles and correction of shading deviations, which is then fused with record attributes to form a twin correlation matrix. Through coupling analysis of shading depth and inspection intervals, and identification of reverse deviations in pipe age and maintenance density, it generates mismatch classification markers. Dynamic rolling scheduling is performed on these mismatch markers. A hazard avoidance confirmation set is obtained through cross-inspector omission rate comparison and accessibility obstacle cluster identification. A graded disposal task group is generated by suppressing false low-risk situations using vacancy-coordinated over-limit difference. Finally, through prediction of sudden changes in on-duty time difference and correction of fatigue coordinate drift, the method integrates scheduling status feedback to output the riser twin inspection results, achieving continuous refinement of the gas riser twin status and proactive risk detection.

The invention provides a road agglomerate fog prediction method based on multi-source data fusion and electronic equipment. The method comprises the following steps: calculating thermal inversion layer structure parameters through a gradient analysis algorithm; calculating dew point temperature sequence data based on the temperature vertical sequence data and the humidity vertical sequence data; on the basis of the humidity vertical sequence data and the dew point temperature sequence data, saturation layer structure parameters are calculated through a saturation criterion algorithm; calculating the overlapping degree of the thermal inversion layer and the saturation layer in the vertical direction; and constructing vector features by using the thermal inversion layer structure parameters, the saturation layer structure parameters, the overlapping degree, the near-surface meteorological data and the underlying surface data, and inputting the vector features into a trained agglomerate fog prediction model to obtain an agglomerate fog occurrence probability. According to the method, through space-air-ground multi-source data fusion, the vertical overlapping degree of a thermal inversion layer and a saturation layer which are closely related to formation of agglomerate fog along a highway is quantified, the formation and dissipation process of the agglomerate fog is reflected, and through deep learning model learning, prediction of the occurrence probability of the agglomerate fog is carried out.

The invention discloses a cobalt ore prospecting method and system based on spectral absorption depth, and belongs to the new technical field of spectral prospecting methods and mineral product prediction. The method comprises the following steps: acquiring short-wave infrared hyperspectral data of a to-be-detected area to-be-selected research object through an image sensor and an intelligent sensing system, performing continuous spectrum removal processing, and extracting mechanism sensitive wave band absorption band depth parameters; constructing a cobalt ore spectrum diagnosis index based on the parameters, and performing inversion on the cobalt element enrichment intensity to form a continuous response field; introducing the response field into a space potential field model to carry out gradient analysis, and revealing enrichment abnormal space convergence characteristics; according to the method, multi-source response information is fused in combination with local space continuity constraints, intelligent identification, prediction, accurate positioning and quantitative exploration of cobalt mineralization abnormity under a complex geological background are realized, and the problem that a traditional remote sensing method is insufficient in ability of identifying concomitant class isomorphic hidden mineralization is solved; the method has the advantages of high recognition degree, accurate positioning prediction, accurate quantitative prediction, wide applicability and the like.

The application discloses a pest prevention method, system and device based on computer vision, and the method comprises the following steps: acquiring a plant X-ray image, performing texture extraction to obtain a plant texture image, and obtaining a pest mask image through gradient analysis; identifying specific position information of a specific type of pest, obtaining a specific type of pest conversion mask image based on the specific position information and the pest mask image, and performing distance conversion; acquiring the intersection of the image boundary of the specific type of pest conversion mask image and the pest area, taking the intersection as a current guide point, obtaining a next guide point through a neighborhood pixel point set of the current guide point, performing iterative analysis to obtain a next guide point set; calculating the distance between all guide points, obtaining the length of a pest body, analyzing the instar, determining the instar of the pest, and preventing the pest through the instar of the pest. The application can analyze the plant X-ray image, obtain the length of the pest body, and realize pest prevention through instar analysis.

The application discloses a groundwater level space-time dynamic monitoring and visual analysis method and system, relates to the technical field of groundwater resource monitoring and visualization, and comprises the following six steps: multi-source data fusion preprocessing, mixed space interpolation of the Kriging and machine learning, isopachous surface generation and optimization, time sequence animation synthesis, abnormality identification and classification, and early warning heat map visualization. The system comprises a data acquisition module, a space interpolation module, an isopachous surface generation module, a time sequence animation module, an abnormality identification module and an early warning visualization module. The application improves the space interpolation precision through multi-source data fusion and mixed interpolation of the Kriging and machine learning, directly presents the space-time evolution of the water level through the isopachous surface and the time sequence animation, identifies abnormal areas such as over-exploitation funnel through space-time gradient analysis, and assists in management decision through the early warning heat map, so as to solve the technical problems of sparse monitoring sites and the difficulty of directly presenting the space-time evolution law in traditional reports.

The invention relates to the technical field of underground drilling and pressure relief, and particularly discloses an irregular hole pressure relief forming method and system. The method comprises the steps that stress baseline data and monitoring point information are acquired, and a time baseline is generated through preprocessing, calibration and time synchronization; generating a hole shape strategy map through regional aggregation and gradient analysis based on the time base line and the stress data; trajectory planning, parameter calculation and feedforward prediction are carried out according to the parameters; a trigger strategy and a staged expansion scheme are formulated; correction parameters are formed through online control, attitude calculation and parameter correction; and finally completing pore-forming execution and structured filing. According to the method, closed-loop control from stress sensing to pore-forming execution is achieved, and the forming precision and the process stability of irregular pores under the complex surrounding rock condition are improved.

This application belongs to the field of environmental engineering and water pollution control technology, specifically relating to an optimized layout method for low-impact development measures that integrates dynamic landscape hydrology theory. The method includes constructing a hydrological connectivity-land use coupling index to dynamically quantify the pollution transport potential of the landscape; objectively identifying the ecological critical threshold of this index based on machine learning and gradient analysis; establishing a multi-objective optimization model with hydrological benefits and costs as objectives and the threshold as a safety constraint; dynamically simulating the underlying surface and hydrological connectivity after implementation for each candidate scheme in the optimization solution, recalculating the index value to verify whether the constraints are met; and finally outputting the Pareto optimal solution set that satisfies the constraints as the optimized layout scheme. This application achieves positive effects in improving the ecological rationality of the scheme, ensuring long-term safety, and realizing automated and efficient decision-making.

The application relates to the technical field of downhole drilling and pressure relief, and particularly discloses a non-regular hole pressure relief forming method and system. The method comprises the following steps: obtaining stress baseline data and monitoring point information, generating a time baseline through preprocessing, calibration and time synchronization; generating a hole shape strategy graph based on the time baseline and stress data through regional aggregation and gradient analysis; planning a trajectory, calculating parameters and feeding forward prediction based on the hole shape strategy graph; formulating a triggering strategy and a phased expansion scheme; forming a correction parameter through online control, attitude solution and parameter correction; and finally completing hole forming execution and structured archiving. The application realizes closed-loop control from stress sensing to hole forming execution, and improves the forming precision and process stability of the non-regular hole under complex surrounding rock conditions.

This invention discloses a multi-scale analysis method and system for pressure fluctuations in a rotating flow field using phase-driven fusion. First, unsteady pressure data of the rotating flow field is acquired and a spatiotemporal snapshot matrix is ​​constructed. Then, the dominant frequency and low-order coherent modes of the pressure fluctuations are extracted through dynamic mode decomposition. Subsequently, a global phase field of the entire flow field is constructed based on the temporal information of the modes. Finally, spatial phase gradient analysis is performed on the phase field to determine the propagation direction of the pressure wave. The global multi-scale phase field is traversed to identify characteristic points of phase changes, and the characteristic points of all dominant frequencies are spatially superimposed. Finally, the spatial modes of each characteristic frequency, the global multi-scale phase field, the propagation direction of the pressure wave, and the location of the core excitation source are superimposed and fused. The modes and phase fields at different times are combined according to a time sequence to synthesize a spatiotemporal evolution holographic map of pressure fluctuations in the rotating flow field. This invention achieves precise localization of the pressure fluctuation excitation source and frequency-space-time correlation analysis of pressure fluctuations in the rotating flow field.

The present application belongs to the technical field of vehicle electromagnetic compatibility detection and fault diagnosis, and particularly relates to a method for electromagnetic radiation dynamic detection and interference source positioning of an electric tractor suitable for plowing operation environment, which overcomes the defect of traditional darkroom test disengaging from actual working conditions, can truly reflect the electromagnetic radiation characteristics of the electric tractor in complex farmland environment, realizes accurate spatial positioning of the interference source, solves the technical problem that the traditional method is difficult to trace the physical position of the interference source, and specifically, the present application overcomes the defect of traditional darkroom test disengaging from actual working conditions through dynamic detection in real plowing operation environment, realizes accurate spatial positioning of the interference source by constructing motion trajectory and reconstructing spatial radiation field, combines gradient analysis and inversion algorithm, locates the physical position of the interference source by correlating and analyzing the radiation anomaly and vehicle operation working condition, and provides support for electromagnetic compatibility engineering rectification.

The application discloses a caterpillar beam weld seam track fitting method based on an improved RANSAC algorithm and particularly relates to the technical field of weld seam track identification and automatic welding path planning. Standardized point cloud data is acquired. Welding spatter points are removed by combining the normal change rate of points with the extraction of weld protruding feature points through local height gradient analysis, a multi-scale point set structure of an overall scale point set and a local scale point set is constructed, a random consistency sampling is performed on the overall scale point set to fit a weld overall trend model, and a local track model set is generated on the local scale point set with the inner point set as a constraint. A track connection graph structure is constructed according to the local curvature change value and the direction continuity score, the screening and splicing of the local track model are completed, and a weld compensation track model is obtained. The compensation track model is subjected to continuous curve fitting to generate a weld center track curve. The application can realize high-precision weld seam track fitting in the presence of welding spatter and local deformation interference.

The invention provides an underground cavern partition stability early warning method based on micro-seismic monitoring. The method comprises the following steps: performing model segmentation on three-dimensional model data of an underground cavern by using a geological condition segmentation rule; performing gradient analysis of longitudinal rock stratum strength and axial fracture density on the N sub-models, and establishing rock stratum transition characteristic data; after the space coordinates of the preset protection object are read, the surrounding rock types of the N sub-models are dynamically recognized based on the rock stratum transition feature data, matching mapping is conducted on the recognition result and the space coordinates, and a surrounding rock risk zone of the preset protection object is configured; data acquisition is executed in real time by using a micro-seismic monitoring network; and inputting the geological parameter set into the corresponding N sub-models, and executing early warning identification according to the surrounding rock risk division. According to the invention, the technical problem that the stability risk of the underground cavern is difficult to accurately identify in the prior art can be solved, and the technical effect of improving the stability risk identification efficiency is achieved.

This invention discloses a dynamic assessment and early warning method for vegetation indices in watershed ecological security monitoring. The method includes: acquiring multi-temporal remote sensing images and surface hydrological and meteorological observation data; performing seasonal baseline separation on the vegetation index time series to accurately identify vegetation degradation and recovery trend segments; combining hydrothermal stress coupling index analysis to quantify the response delay and sensitivity characteristics of vegetation to environmental changes; constructing a dynamic assessment model based on stress response parameters; and identifying ecologically vulnerable transition zones through spatial gradient analysis to generate tiered early warning signals. This invention overcomes the spatiotemporal limitations of traditional monitoring, accurately predicts potential ecological risks, and provides an early intervention window for watershed management departments.

The present application relates to the technical field of crested ibis protection, and discloses an intelligent evaluation system for crested ibis habitat environment. The water quality dynamic evaluation module of the system acquires water quality sensor data, adjusts the proportion of acid-base balancing agent and mineral supplement; the vegetation coverage gradient analysis module compares the vegetation coverage density change rate with the crested ibis habitat demand based on the water quality regulation result and adjusts the repair rate; the soil moisture regulation module calculates the soil moisture content of multiple sampling points according to the vegetation adjustment result and adjusts the water management strategy; the meteorological change prediction and regulation module acquires meteorological data, calculates the future humidity change and adjusts the proportion of water and mineral supplement; the ecological factor linkage optimization module extracts the crested ibis activity frequency and food source data, calculates the diurnal ecological factor supply proportion and adjusts the management time sequence, and finally generates the intelligent evaluation result of the crested ibis habitat environment, so as to realize the accurate and systematic regulation of the crested ibis habitat environment.

The invention relates to the technical field of geological disaster monitoring, in particular to a rock collapse accumulated debris rheological property prediction system and method based on deep learning. The method comprises the following steps: acquiring topographic data of a rock collapse accumulation area; performing gradient analysis by using topographic data, and marking a high-gradient region; performing rock collapse debris detection based on the high-gradient region to obtain rock collapse debris data; inputting the rock collapse debris data into a pre-trained deep learning model, and performing debris liquefaction potential analysis to obtain liquefaction potential data; performing slurry forming simulation according to the liquefaction potential data to obtain slurry forming evolution data; performing rheological initial parameter calculation based on the pulping evolution data to obtain rheological parameter data; performing debris slurry flow simulation according to the rheological parameter data to obtain slurry flow data; and identifying shear thinning characteristics according to the slurry flow data. According to the method, the accuracy of rock collapse debris liquefaction potential identification and rheological property prediction is improved based on a geological disaster monitoring technology.

The invention discloses a photovoltaic string fault diagnosis method based on an improved rotating forest algorithm, and relates to the technical field of fault diagnosis. The method comprises the steps that string operation basic data is acquired and preprocessed, and a first feature data set is constructed; marking based on the state characteristics of the inverter and the combiner box to generate a first label; performing gradient analysis on the string time sequence characteristics to generate a second label; performing benchmarking analysis on the same type of group strings in the same upper layer, and generating a third label based on the deviation; acquiring defect features based on the component edge AI to generate a fourth label; and constructing a gating loop optimization rotating forest model, performing aggregation diagnosis on the data sets corresponding to the four types of labels, and outputting a string fault result. The technical problems that an existing photovoltaic power station equipment fault diagnosis model is low in robustness and has many invalid alarms are solved, and the technical effects that aggregation diagnosis is carried out by introducing the multi-source feature tags and fusing the gating loop optimization rotating forest model, and the accuracy and robustness of photovoltaic string fault recognition are improved are achieved.