279 results about "Spatial heterogeneity" patented technology

The invention relates to a regional distributed photovoltaic power prediction method and system based on multivariate data cross-modal fusion, and belongs to the technical field of photovoltaic power prediction. Firstly, adaptive extraction of meteorological data spatial features and power data spatial heterogeneity is realized through a dynamic perception convolution kernel of AG-CNN, and a high-quality spatial basis is provided for cross-modal fusion; secondly, utilizing a two-dimensional dynamic attention mechanism of a DR-Transform to deeply mine cross-modal association of'historical power time sequence-future weather driving ', adapting to fusion requirements in different scenes through dynamic weight distribution, and capturing association evolution in a long period; thirdly, designing a cross-modal feature bridging module; and finally, verifying the robustness of the model in a typical scene, and ensuring that the model meets the precision requirement of multi-scale scheduling of the power grid.

The invention discloses a wetland carbon sink benefit evaluation method and system based on remote sensing data, and relates to the technical field of carbon emission evaluation. A wetland carbon sink benefit evaluation system based on remote sensing data comprises a carbon sink evaluation zoning module and a carbon sink benefit evaluation module. According to the invention, through multi-modal remote sensing data fusion and fine grid division, high-resolution analysis of spatial heterogeneity of a wetland region is realized, different types of carbon sink regions and distribution boundaries thereof can be accurately identified, and scientificity and particles of carbon sink partitioning are improved; a dynamic index characteristic graph layer is constructed through a remote sensing time sequence, a hydrological pulse response factor graph layer is constructed in combination with hydrological data, and clustering division is performed on the hydrological pulse response factor graph layer, so that dynamic identification and pixel level benefit calculation of a dynamic carbon sink area are realized.

The invention discloses a black soil area ecological stability-restoring force coupling zoning method and system, and relates to the technical field of ecological environment protection. Collecting multi-source spatio-temporal data of a target black soil area, and analyzing and obtaining the ecological stability and ecological restorability of the area; according to the spatial heterogeneity of the ecological stability and the restoring force, spatial difference characteristics of ecological variables in different geographic units are mined, and spatial coupling analysis is carried out on the ecological stability and the ecological restoring force through restoring force prediction; and according to a coupling analysis result, performing ecological stability-restoring force coupling partition management on the target black soil area. According to the invention, space identification and differentiation management of the ecological system state of the black soil area are realized.

The invention discloses a tidal flat automatic extraction method fused with an irregular hexagonal grid. The method comprises the following steps: step 1, constructing a time sequence remote sensing data set; step 2, establishing a precision verification sample library; step 3, screening spectral indexes; 4, synthesizing an extreme tide level image; 5, generating an irregular hexagonal grid; step 6, carrying out adaptive threshold segmentation; step 7, secondary elimination of confused ground features; step 8, tidal flat range extraction and precision verification; according to the method, based on a time sequence remote sensing image set, pixels at all positions are screened through a specific spectral index maximum value, a new extreme tide level image is formed through combination, spatial heterogeneity of a complex coastal zone is adapted through an irregular hexagonal grid and a bottom-to-top hierarchical merging strategy, local self-adaptive threshold segmentation is achieved in cooperation with a local Otsu algorithm, and a new extreme tide level image is obtained. The optimal grid scale is automatically determined by means of an elbow method, and the confused ground features are secondarily eliminated through the time sequence standard deviation, so that the problem of miscarriage of the confused ground features is effectively solved, and high-precision full-automatic extraction of the tidal flat is realized.

The invention discloses a distributed hydrological model parameter two-stage calibration method based on remote sensing data and actually measured flow. The method comprises the following steps: collecting and preprocessing basic data required for constructing a distributed hydrological model, and remote sensing data and actually measured flow data for calibrating model parameters; constructing a distributed hydrological model, determining hydrological parameters needing to be calibrated in the model, and setting a value range of each parameter; performing grid-by-grid parameter calibration by utilizing remote sensing data to obtain model parameters with spatial heterogeneity characteristics; performing optimization adjustment on grid-by-grid model parameters based on remote sensing data calibration by utilizing actually measured flow data; and finally selecting a parameter combination with the optimal simulation precision as a finally determined model parameter for researching the drainage basin. Spatial distribution information of remote sensing data and time sequence precision of actually measured flow data are fused, and a multi-target calibration scheme considering parameter spatial heterogeneity and calibration result reliability is constructed.

The invention discloses a mountain torrent disaster danger area dynamic demarcation method based on multi-source data coupling, and the method comprises the steps: obtaining meteorological radar rainfall forecast data, surface deformation monitoring data and high-precision DEM topographic data in real time, carrying out the data preprocessing, and generating a fused multi-source data set; constructing a distributed hydrological model, and calculating the production and convergence flow of the sub-basins; training a Seq2Seq-LSTM model on the basis of rainfall-water level-swelling speed time sequence data, predicting danger areas in a future set time period, and dynamically dividing high / medium / low risk areas by adopting an SOM-CLARA clustering algorithm and taking rainfall intensity, earth surface deformation rate and gradient as feature vectors; and determining the early warning level of each risk area in combination with the critical rainfall threshold and the spatial heterogeneity index. Through the multi-source data dynamic coupling and self-adaptive demarcation algorithm, the dangerous area range is dynamically adjusted, and the accuracy and timeliness of mountain torrent disaster risk prevention and control are remarkably improved.

The invention discloses an urban inland inundation prediction method and system based on space attention and depth fusion, and relates to the technical field of urban inland inundation prediction, and the method comprises the steps: laying monitoring points in a waterlogging-prone risk area, and constructing a dense perception network; constructing a dynamic space attention network, quantifying rainfall-ponding space correlation, calculating a dynamic influence weight through a learnable weight matrix, and generating an attention weight matrix to represent the influence of rainfall points on ponding points; training a long-short-term memory network model, and fusing the feature representation of the spatial correlation information and a historical ponding depth sequence to carry out ponding depth time sequence prediction; and based on topographic feature extraction, physical guide diffusion and point time sequence prediction data conversion, expanding topographic-guided point-surface data, and generating a surface region ponding risk map. Through fusion of a spatial attention mechanism and deep learning, rainfall-ponding spatial heterogeneity association is effectively captured, the accuracy and timeliness of urban inland inundation prediction are improved, and scientific decision support is provided for urban flood control and disaster reduction.

The invention discloses a rock slope multi-mode instability probability analysis and risk evaluation method based on a drilling constraint condition random field, and belongs to the technical field of geotechnical engineering and engineering geology. According to the method, the conditional random field which is strictly consistent with the actually measured data is constructed under the condition of limited drilling exploration data, so that the reasonable expression of the spatial heterogeneity of the rock mass mechanical parameters is realized, and the authenticity of a numerical model and the reliability of an analysis result are greatly improved; the defect of insufficient result credibility caused by lack of actual measurement constraint of the unconditional random field is effectively overcome; through a large amount of multi-sample numerical simulation, multiple instability modes possibly occurring on the rock slope under different parameter space distribution conditions can be identified, the occurrence probability of each instability mode is quantitatively calculated, various risks possibly existing on the slope can be comprehensively grasped, only the most dangerous single condition is concerned, and the risk assessment accuracy is improved. And the accuracy and engineering applicability of slope stability analysis and risk evaluation are improved.

The invention provides a spatial diagnosis system for mountainous city facility layout. Comprising the following steps: a data acquisition and preprocessing module preprocesses initial multi-source spatial data of a target area to obtain target multi-source spatial data; the space unit division module performs space unit division on the target area to obtain a plurality of space units; a public service facility reachability calculation module determines public service facility reachability of a target demand point in the target area according to the target multi-source spatial data; the spatial difference diagnosis module identifies spatial heterogeneity of each spatial unit on different scales according to the target multi-source spatial data; the problem space positioning module performs problem space positioning according to the public service facility accessibility of the target demand point and the topographic features of the target area; and the optimization scheme generation module generates an optimization scheme for each problem space. Therefore, according to the regional characteristics of the mountain city, optimal configuration schemes of different facilities are provided in a targeted manner, so that the scientificity and rationality of walking facility configuration are improved.

The invention discloses a federated learning prediction method and system for geological risks of a multi-section tunnel, particularly relates to the technical field of tunnel engineering safety monitoring, and is used for solving the problems of feature conflicts and model precision reduction caused by neglecting geologic spatial heterogeneity in existing federated learning. The method comprises the following steps: acquiring geological risk feature vectors and three-dimensional geological coordinates output by a local model of each section; calculating the cosine similarity and the space Euclidean distance of the feature vectors of any two sections; carrying out weighted summation on the cosine similarity and the reciprocal of the space Euclidean distance according to an exponential attenuation weight to generate a geological association weight; generating an event cluster space coherence coefficient based on surrounding rock microseismic event space-time coordinate clustering; when the coefficient is lower than a threshold value, isolating an abnormal section, and weighting and aggregating model parameters of non-isolated sections by adopting a geological association weight to generate a global model; the collaborative modeling of the geological space structure and the dynamic risk is realized, and the risk prediction precision in a multi-section complex geological environment is improved on the premise of ensuring data privacy.

The invention belongs to the technical field of equipment planning, and discloses an automatic planning method for intelligent agricultural Internet of Things equipment, which comprises the following steps: dividing a target region into grid units, collecting soil moisture parameters and nutrient parameters of each grid, and establishing a regional digital twinborn model; obtaining a target crop of each grid, obtaining growth information from the crop knowledge base, and executing time sequence simulation to generate a water and fertilizer demand table of each grid; executing a water-fertilizer balance strategy based on the water-fertilizer demand table, and generating a water-fertilizer supply plan of each grid; analyzing spatial distribution characteristics of supply plans, and clustering adjacent grids with similar plans into irrigation management partitions; and generating a pipe network topological structure and an equipment planning scheme based on the partition layout. According to the method, a traditional uniformly distributed area is scientifically divided into management subareas with consistent internal requirements, so that unified management of the same subarea is realized, and the problem of supply and demand mismatching caused by neglect of spatial heterogeneity in a traditional method is solved.

The invention discloses a regional consumption demand calculation method based on e-commerce platform data, and belongs to the technical field of regional consumption demand calculation. The problem that a regional consumption demand calculation method is inaccurate in calculation is solved. The method comprises the steps of collecting basic data, obtaining preprocessed regional transaction data through an abnormal value self-adaptive suppression and neighborhood spatial interpolation method, carrying out spatial association, and calculating a regional commodity consumption characteristic entropy value; decomposing a time trend and a space trend in the consumption data through a periodic function and a space overflow model, and calculating a decomposed consumption trend and a decomposed daily average consumption trend; considering the influence of price elasticity and logistics timeliness, and calculating a consumption demand coupling factor; performing spatial heterogeneity correction through a geographical weighted kernel function to obtain a corrected consumption demand coupling factor; the regional commodity demand index is calculated through nonlinear normalization and supply constraint. Compared with a traditional theoretical demand calculation method, the method has the advantage that the landing conversion rate is improved.

The invention provides a seepage risk dynamic quantification method and device, and relates to the technical field of engineering prediction.The method comprises the steps that seepage parameter observation data of different observation points in a target area and geological environment factor data related to seepage parameters are obtained; constructing a multi-scale geographically weighted regression model based on the permeability parameter observation data and the geological environment factor data to obtain a deterministic space trend field of permeability parameters; calculating a spatial residual error of the penetration parameter observation value relative to the deterministic spatial trend field, and performing Kriging interpolation on the spatial residual error to obtain a random residual error field and uncertainty information; fusing the deterministic space trend field and the random residual field to obtain a permeability parameter space distribution field; and based on the seepage parameter space distribution field and the uncertainty information, utilizing a Bayesian assimilation algorithm to obtain a seepage risk dynamic quantification result. According to the method, the core defects of insufficient description of seepage parameter spatial heterogeneity and single risk prediction static state in a traditional method can be solved.

The invention provides a mine ecological restoration monitoring system based on unmanned aerial vehicle inspection, and the system comprises the following functional modules: a data fusion module which periodically collects visible light and multispectral images of a mining area, and generates a high-resolution data base map through fusion processing; the ground feature classification and diversity quantification module is used for completing ground feature classification by utilizing a random forest algorithm based on the data base map and calculating a Shannon diversity index through a space grid; the dynamic diagnosis and regulation and control strategy generation module is used for calculating a time gradient and a spatial heterogeneity change rate according to the multi-period Shannon diversity index, diagnosing an ecological state and generating a regulation and control strategy according to the time gradient and the spatial heterogeneity change rate; and the ecological toughness evaluation and early warning module is used for calculating an ecological toughness index by fusing the indexes and realizing graded evaluation and early warning of the ecological restoration situation. According to the invention, dynamic quantitative monitoring of the evolution process of the mine ecosystem is realized, the ecological change trend can be accurately identified, and graded early warning and regulation decision support is provided for restoration engineering.

The invention discloses a multi-component gas parameter storage control method and system, and belongs to the technical field of automatic control, and the method comprises the steps: collecting initial environment parameters, obtaining a stacking form image of a stored material, recognizing stacking feature parameters, constructing a gradient pre-judgment model to obtain a gradient pre-judgment interval, and marking a high-risk region; constructing a layered node monitoring network, collecting local parameters of each monitoring node in real time, and identifying an abnormal deviation region in combination with a total environment mean value; coupling correlation is constructed, predictive gradient intervention is carried out, an optimal airflow path is calculated, a gas supply strategy is adjusted, a dynamic impedance adaptation strategy is utilized, a high-impedance area is identified, multi-target optimization is utilized, differential regulation is carried out, and a targeted regulation strategy is generated; and setting an update time interval and calculating a pre-judgment deviation rate to carry out deviation correction, thereby ensuring that local parameters are stabilized in a safe range, avoiding the out-of-control problem caused by spatial heterogeneity and dynamic coupling, and ensuring the stability of the whole storage effect.

The present application relates to the technical field of geological disaster risk assessment, and discloses a landslide-prone prediction method based on multi-scale geographic weighted regression and spatial heterogeneity partitioning, which solves the problems of low local accuracy, poor area efficiency and insufficient prediction stability caused by single spatial scale, uniform modeling and single evaluation system in traditional landslide prediction models. The present application organically combines multi-scale geographic weighted regression with spatial heterogeneity partitioning, first adaptively assigns a dedicated optimal bandwidth to each disaster-inducing factor through a multi-scale regression model to accurately depict the spatial non-stationary characteristics of landslide disaster-inducing relationships, then reconstructs the discrete regression coefficients into a spatial continuous surface with the help of Kriging interpolation, dynamically extracts the dominant factors with the optimal discrete and natural discontinuity method to realize the heterogeneity partitioning of the study area, takes each partition as an independent unit to carry out landslide-prone deduction, and constructs a multi-dimensional comprehensive evaluation system from the dimensions of prediction accuracy, area rationality and uncertainty robustness.

This invention discloses an image recognition-based method for assessing grape powdery mildew. By introducing standardized sample preparation and uniform inoculation methods, this invention reduces the interference of randomness and spatial heterogeneity of natural field infection on the assessment results, making the disease occurrence process more controllable among different materials, thereby improving the stability and comparability of phenotypic data. Simultaneously, image recognition and machine learning technologies are used to automatically identify and calculate the area of ​​powdery mildew-infected regions, avoiding the reliance on the experience of assessors in traditional manual visual grading. This enables a continuous quantitative description of the degree of powdery mildew infection, improving the consistency and accuracy of the assessment results. This invention constructs a large-scale image analysis workflow that can be used for batch processing, significantly improving the efficiency of disease assessment, reducing the workload of manual investigation, and making this method applicable to applications requiring the processing of large numbers of samples, such as germplasm resource evaluation and breeding material screening.

The invention relates to a soil heavy metal risk assessment method for a fuzzy region, fully considers spatial heterogeneity of soil heavy metals and a complex relationship between the soil heavy metals and agricultural products, and provides a fuzzy region averaging method, an improved potential ecological risk index method and an agricultural product risk index method. Therefore, soil heavy metal risk assessment of the fuzzy region is realized, and technical reference is provided for regional soil heavy metal risk assessment.

The invention relates to the technical field of environmental engineering, in particular to a nitrogen and phosphorus pollution load analysis method and system for large-range drainage basin water quality. The method comprises the following steps: acquiring a currently detected soil sub-region and a currently detected water sub-region; according to the nitrogen and phosphorus content collaborative fluctuation degree detected before and the neighborhood nitrogen and phosphorus content fluctuation degree of the pixels in the soil sub-region, the pollution accommodation degree is obtained; according to the difference between the nitrogen and phosphorus diffusion directions and the difference between the nitrogen and phosphorus content fluctuation degrees of any two soil sub-regions, combining the soil sub-regions to obtain a pollution migration region, and determining a nitrogen and phosphorus pollution migration path; and according to the pollution accommodation degree of the soil sub-region in the pollution migration region where the soil sub-region is located, the consistency degree of the nitrogen and phosphorus migration direction and the nitrogen and phosphorus attenuation degree on the nitrogen and phosphorus pollution migration path, the water body pollution influence degree is obtained, and the water body pollution load value is obtained in combination with the nitrogen and phosphorus content. Spatial heterogeneity and dynamic accommodation and transmission effects are integrated, and the accuracy of nitrogen and phosphorus pollution load analysis is improved.

The application discloses a kind of based on spectrum-process-terrain coordination's farmland low-lying quick extraction method of waterlogging, comprising the following steps: step one, satellite remote sensing image acquisition and pretreatment;Step two, waterlogging area grading mark;Step three, process quantity extraction;Step four, red edge area index calculation;Step five, terrain factor processing and feature combination;Step six, model training and optimal feature subset screening;The application is by constructing bare soil period soil moisture variation process quantity and growth period crop growth change process quantity, accurately depict the evolution track of waterlogging with growth stage, break through the limitation that single phase cannot reflect process difference;By constructing red edge area index, enhance the spectrum difference of waterlogging, distinguish bare soil period easily confused ground class, avoid growth period vegetation canopy shielding interference;Through terrain factor, reflect the influence of topography on water depth, holding water time length, improve the identification accuracy of waterlogging spatial heterogeneity;By using random forest algorithm, improve the accuracy of waterlogging grading.

The invention relates to the technical field of digital soil mapping, and particularly discloses a soil organic carbon global and local integrated modeling method considering spatial heterogeneity. According to the method, global linear, global nonlinear and local spatial heterogeneity relationships are respectively captured through a mixed linear model, a support vector machine and a multi-scale geographically weighted regression model, a KAN-Transform architecture is utilized to generate a spatial adaptive integrated weight, and Bayesian optimization is combined to improve prediction precision. The problem that a traditional single model or integration method is insufficient in depicting complex spatial relations is effectively solved.

The present application relates to a dynamic regulation method for probiotic solid-state fermentation of oil tea chaff based on digital twinning, and belongs to the technical field of probiotic solid-state fermentation. The method comprises the following steps: constructing a multi-agent decision model based on a reinforcement learning algorithm, learning the multi-agent decision model, using the multi-agent decision model to optimize control parameters, and controlling the working environment of the probiotic solid-state fermentation tank according to the optimal working parameters. The real-time oil tea chaff fermentation data are compared with the oil tea chaff fermentation data within the preset time, and online optimization is performed according to the comparison result. The present application can deeply integrate multi-scale process mechanism, real-time spatial heterogeneity perception, and dynamic regulation method with self-adaptive collaborative decision-making capability to overcome the defects of the prior art, such as regulation lag, neglect of spatial differences, insufficient model prediction accuracy, and limited optimization decision-making capability, thereby realizing efficient, stable and high-quality production of the probiotic solid-state fermentation of oil tea chaff.

The application discloses a forest structure and intelligent coupling analysis and modeling method for absorbing photosynthetically active radiation, and belongs to the technical field of forest ecology remote sensing, structure analysis and computer simulation. The method focuses on fusing airborne and handheld laser radar information, accurately inverting forest structure indexes, covering single-tree level attributes, neighborhood spatial structure indexes, gap structure, stand density SD and slope SL; three-dimensional radiation transmission simulation is carried out; through the construction of a single-tree three-dimensional model library and a digital sample plot, high-precision three-dimensional forest reconstruction and hierarchical APAR simulation are realized, and photosynthetic spatial heterogeneity parameters are extracted; Spearman correlation analysis is carried out based on the structure and APAR parameters, and PCA dimension reduction and KMeans clustering are combined to identify the dominant structure mode and spatial type; based on the clustering result, a typed structure-APAR coupling integrated learning model is constructed, and the regulation mechanism of single-tree attributes and neighborhood structure on APAR is revealed.

This invention belongs to the field of fracture characterization and reservoir hydraulic fracturing simulation technology. It discloses a method and related equipment for characterizing natural fractures and simulating reservoir damage evolution using a phase-field model. By distinguishing primary and secondary natural fractures based on a discrete fracture network and constructing a model, it achieves accurate characterization of the different sizes and spatial heterogeneity of primary / secondary natural fractures in the model. Natural fractures are characterized as mechanically weak surfaces with differentiated parameters. Hydraulic fractures can capture and interact with natural fractures. Combining a phase-field damage model and a rock mechanics constitutive model, it can accurately simulate key behaviors such as activation and deflection during natural fracture propagation without pre-setting fracture paths, avoiding numerical instability and low computational efficiency. Through dynamic coupling analysis of the entire fluid-solid-damage process, it takes into account the influence of multiple factors such as geostress and fluid pressure, improving the accuracy of phase-field simulation, reducing simulation distortion, providing reliable support for fracturing design optimization, helping to improve reservoir stimulation effects and production benefits, and ensuring the smooth implementation of fracturing processes.

The invention belongs to the technical field of marine environment monitoring and artificial intelligence, and discloses a method for intelligently monitoring coastal water eutrophication. The method comprises the following steps: acquiring and matching spectrum, space and time parameters of a remote sensing image and actually measured data; constructing a multi-modal feature data set; a hybrid prediction model formed by cascading a Transform network and a support vector regression model is constructed and trained, and the Transform network performs unified modeling and deeply fuses multi-source features such as spectrum, space and time through a self-attention mechanism of the Transform network, so that deep intelligent fusion of the multi-source features is realized; performing water quality index prediction and eutrophication state evaluation by using the trained model; and finally, carrying out uncertainty quantization on a prediction result by adopting a Monte Carlo random inactivation technology. According to the invention, the problems of low precision and weak generalization caused by spatial heterogeneity and complex optical characteristics in coastal water monitoring are effectively solved.

The present application relates to the field of geographic information analysis, and particularly relates to a cityization influence threshold identification method and device for supply and demand of coastal wetland ecological function, based on spatial position coordinate data of a plurality of grid units in a coastal area, a cityization level comprehensive index, and a supply and demand ratio of comprehensive ecological function of the coastal wetland, a preset initial spatial weighted threshold identification model is fitted to identify a critical point of the cityization level on the supply and demand relationship change of the ecological function of the coastal wetland, to reflect the stage response characteristics of the supply and demand of the ecological function of the coastal wetland to the cityization and the spatial heterogeneity characteristics of the cityization influence mechanism in different geographic units, to effectively identify the spatial difference threshold characteristics of the supply and demand relationship of the ecological function of the coastal wetland in the response process to the cityization, and to provide a refined and partitioned scientific basis for regional ecological management strategies.

The invention discloses an urban residence land price evaluation method considering data imbalance and spatial heterogeneity. The method comprises the steps of collecting historical residence land sample data of a research area and performing spatial position matching; performing hierarchical index evaluation processing on the historical residential land sample data by using a multi-hierarchical index system, and performing fusion arrangement to obtain a residential hierarchical index land price data set; the land price evaluation combination model carries out model training of residence land price prediction by using a residence level index land price data set, and a differential evolution algorithm model DE dynamically searches a hyper-parameter combination of an XGBoost model and carries out optimization processing; and obtaining residential land multi-source data of the research block to obtain indexes of all levels, inputting the indexes into the land price evaluation combination model, and outputting and obtaining a residential land price evaluation prediction result. According to the method, the double problems of unbalanced sample price data and insufficient spatial heterogeneity identification are solved, high-precision urban residential land price evaluation is realized, and the accuracy and automation level of land price evaluation are improved.

This invention discloses a method for constructing background values ​​of river and lake sediments based on curvature extrema and dynamic zoning, relating to the field of environmental benchmarking and ecological risk assessment technology. The method includes the following steps: zonal sampling and multi-element determination; construction of a background sediment database; construction of zonal background values; calculation of the entire lake's background value; and three-way verification of the background value. This invention verifies the natural background values ​​of heavy metals in river and lake sediments through dynamic zonal sampling, geometric feature analysis, and multi-source verification. It provides a dynamic zonal sampling strategy that can truly reflect the spatial heterogeneity of river and lake sediments, establishes a background sediment identification system that balances cost and accuracy, and includes both precise approaches based on sophisticated instruments and practical approaches based on empirical knowledge. This enhances the applicability of the method under different conditions. An objective background value calculation model based on data geometric features is introduced to eliminate the biases caused by subjective threshold settings and outlier sensitivity in traditional statistical methods.

The invention discloses a loess mechanical parameter spatio-temporal interpolation prediction model construction method based on big data, and relates to the technical field of geotechnical engineering data analys.The loess mechanical parameter spatio-temporal interpolation prediction model construction method comprises the following steps that loess mechanical parameter data, environmental factor data and geologic structure data are collected through a multi-source sensor, and spatio-temporal unified multi-modal data are constructed; designing a collaborative kernel function fusing spatial heterogeneity and time dynamics on the basis of multi-modal data, and quantifying relevance between mechanical parameters and space-time positions and environmental factors; according to the collaborative kernel function, dynamically optimizing a space weight, a time attenuation coefficient and a multi-source data fusion weight, and generating a parameter-adaptive interpolation model; carrying out distributed training and calculation acceleration on the interpolation model by adopting a block matrix approximation and parallel calculation architecture; through the trained model, a mechanical parameter prediction value of a target space-time position is calculated in real time, and the calculation efficiency is improved through a tensor expansion algorithm.

The application discloses a crop yield statistical data downscaling method based on integrated agricultural condition index standardization deviation from mean, which comprises the following steps: A1: data preparation; A2: agricultural condition parameter inversion; A3: integrated agricultural condition index construction; integrated agricultural condition index of key phenophase is constructed, and the correlation between crop agricultural condition parameters and statistical yield in the key phenophase is analyzed, the integrated agricultural condition index of the key phenophase based on the normalized regression coefficient weight determination method is constructed, and the integrated agricultural condition index of the key phenophase based on the yield correlation is verified in precision; A4: yield statistical data downscaling and precision analysis. In addition to comprehensively considering the correlation between crop agricultural condition parameters in the key growth period and the yield change of crops, the application also considers the influence of spatial heterogeneity between integrated agricultural condition parameters pixels on the yield change of the region at the regional level. The crop growth state spatial heterogeneity expression method of the integrated agricultural condition parameter space-time dynamic weight proposed by the application has important significance for improving the simulation capability of the spatial change of the yield of the crop pixel scale.