18 results about "Proportional Stratified Sampling" patented technology

This invention provides a method for high-speed train bearing fault diagnosis and selection based on multi-model comprehensive evaluation. The method first obtains bearing fault feature vectors, then divides the training and test sets using 8:2 stratified sampling, mitigating sample imbalance bias through a sample weighting strategy. Next, it constructs a diagnostic system comprising five models, including random forest and support vector machine, and employs optimal hyperparameters for parallel training. Subsequently, a multi-dimensional evaluation system for accuracy and efficiency is established, and model reliability is verified using a confusion matrix. Finally, model selection is tailored to specific scenarios: multilayer perceptrons are used for real-time monitoring, gradient boosting trees for high-precision scenarios, and random forests or K-nearest neighbors for lightweight deployment. This achieves scientific multi-model selection, balancing diagnostic accuracy, real-time performance, and stability, adapting to the maintenance needs of high-speed trains, and demonstrating strong engineering practicality.

The application provides a soil monitoring sampling method, system, device and medium, the region to be monitored is divided into grid by spatial position and historical soil information, forming a candidate sampling unit with spatial correlation, in each candidate sampling unit, the physical parameters representing the state of the soil surface and shallow layer are collected, and a high-dimensional feature vector is generated; based on all high-dimensional feature vectors, a dynamic sampling priority scoring model reflecting the degree of spatial difference of soil is constructed, and a target sampling unit is determined; a depth stratified sampling strategy is performed on the target sampling unit, the sampling depth interval is adaptively determined, and then a representative soil sample is obtained; the soil sample is associated with the corresponding spatial position, sampling depth and soil state information, and structured soil sampling data is generated. The scheme of the application can realize the spatial-depth joint optimization of soil sampling, and accurately capture the real state and spatial variation law of the soil environment.

This utility model provides a stratified sampling device for marine environmental surveys, relating to the field of marine environmental sampling equipment. It includes: a base plate with a rectangular notch on the front of its upper end face; a winding mechanism mounted on the rear of the upper end face of the base plate, with a flexible rope wound onto the winding mechanism; and a collection bottle connected to the flexible rope via a C-shaped connector and a fixing bolt. This utility model, through the cooperative design of the anti-loosening component and the locking gear, can lock and limit the locking gear and the fixing bolt, preventing the fixing bolt from loosening under stress, thus improving the stability of the collection bottle on the flexible rope and avoiding affecting the reliability of the stratified sampling device. It solves the problem in existing stratified sampling devices where, although the collection bottle can be fixedly connected to the flexible rope via fixing bolts, the fixing bolts are prone to loosening in dynamic marine environments, leading to a decrease in the stability of the collection bottle on the flexible rope and affecting the reliability of the sampling stratified sampling device.

This utility model discloses a soil sampling device for ecological restoration with stratified sampling capability, relating to the field of soil sampling technology. It includes: a main body of the sampling device, with a rotatable shaft inside; a sampling ring fixedly installed at the bottom of a sampling cylinder, with an opening on one side; the sampling ring having a V-shaped cross-section; and a sampling basin rotatably installed on the top of a plug and located inside the sampling cylinder, with the top of the sampling basin's center fixedly connected to the bottom of the shaft. This utility model, through the rotation of the shaft, drives the sampling basin, partition, and scraper to move. The scraper moves, scraping soil from the opening of the sampling ring into the sampling basin. The partition separates each sampled soil, thus achieving the effect of convenient and efficient multiple sampling of hard soil, avoiding substandard sampling due to excessively hard soil, thus avoiding multiple samplings and wasting labor, and improving sampling efficiency.

The utility model provides a kind of hydraulic engineering geological survey sampling device, it is related to hydrogeological engineering survey technical field, including cylinder, by coaxial setting upper sampling cavity, middle sampling cavity and lower sampling cavity constitute three-dimensional sampling space;Negative pressure source, it is communicated to each sampling cavity, directional negative pressure is formed when sampling.The utility model can realize high-precision stratified sampling without cross-contamination, by orthogonal distribution upper, middle and lower water inlet channel, make the water body of different depth in geological survey stationary surface water along independent path into corresponding sampling cavity, completely avoid interlayer mixing from physical structure, ensure stratified data authenticity;The utility model can optimize fluid flow, guarantee sample integrity and representativeness, the horn-shaped water inlet of anti-turbulence cover reduces flow velocity, combined with spiral guide flow channel, convert turbulent flow into stable laminar flow, reduce bubble disturbance and particulate matter settlement, significantly improve the originality of sample, meet high-precision water quality analysis demand.

This invention discloses a literature screening method, system, device, and storage medium based on multi-language model stability determination, relating to the fields of information technology and literature processing technology. The method constructs a complete literature screening quality assessment system through core steps such as data import, preprocessing, dual-model multi-round screening, stability determination and early stopping, arbitration judgment, and stratified sampling evaluation. The system adopts a multi-module collaborative architecture, achieving a technological improvement from "single judgment" to "quantifiable stability judgment" through prompt word rotation, quantifiable stability determination, and intelligent arbitration mechanisms. This invention effectively solves key technical problems in multi-source literature data processing, significantly improves the accuracy, stability, and traceability of screening results, and provides an efficient and reliable solution for scenarios such as system reviews and scientific and technological intelligence analysis.

This utility model discloses a high-efficiency water and soil sample collection device, including a sampling bucket, a stratified sampling component, and an isolation component. The stratified sampling component includes a rotating cover plate, which is movably connected to the sampling bucket. A pull rope is fixedly connected to the top of the rotating cover plate. A support plate is fixedly connected to the top of the sampling bucket. A guide frame is fixedly connected inside the support plate. A guide wheel is movably connected inside the guide frame. The pull rope passes through the guide frame and is movably connected to the guide wheel. A handle is fixedly connected to the surface of the sampling bucket, and a traction rope is fixedly connected to the surface of the handle. The traction rope is located at the top of the sampling bucket. This utility model utilizes the stratified sampling component for stratified sampling, enabling sampling of different water layers. The sampling results are diverse, and the sampling target can be achieved without multiple samplings.

This invention discloses a water quality monitoring stratified sampling device, comprising a sampling bucket, a support rod at the top of the sampling bucket, an air rod inserted through the middle of the support rod, the air rod penetrating the sampling bucket, a piston plate at the bottom of the air rod, an exhaust channel on one side of the piston plate, and an opening and closing part within the exhaust channel; a locking rod on one side of the support rod, the outer end of the locking rod locking into a lock hole on the top side of the sampling bucket, a rope on the outer side of the locking rod, and a float at the other end of the rope; a main water inlet at the center of the bottom of the sampling bucket, with a valve on the main water inlet; several water outlets on one side of the sampling bucket, with water outlet pipes connected to the water outlets; several rotating parts inside the sampling bucket, dividing the interior of the sampling bucket into several sampling chambers, with an auxiliary drive part at one end of each rotating part, which is installed on the outer wall of the sampling bucket, to achieve independent segmentation of water samples at different depths, effectively avoiding cross-contamination between water samples and improving the accuracy of monitoring data.

This invention provides a stratified sampling method, system, and storage medium for soil remediation. The method includes: integrating historical knowledge base and field survey data; generating a three-dimensional posterior probability distribution and uncertainty field of pollutants using a Bayesian model; generating sampling cells using non-uniform Voronoi partitioning based on the uncertainty intensity plane map; automatically identifying key candidate sampling depths on the vertical probability profile of each cell by analyzing local extrema and curvature changes; calculating priority scores by comprehensively considering the uncertainty, probability value, and derivative amplitude of each depth; and selecting the optimal sampling points after global sorting.

This invention provides a method and system for optimizing the deployment of nearshore fisheries resource monitoring stations and navigation routes based on multiple objectives. The invention first determines the optimization objectives, then selects the optimal coupling model through evaluation indicators, and obtains the true distribution data of resource density through the optimal coupling model. It employs the Delphi method combined with information entropy and coefficient of variation as dual objective weights, along with a target-oriented correction coefficient, to obtain the dynamic weights of the optimization objectives. Three-level dynamic stratified sampling is performed using a heterogeneity threshold, and the optimal number of stations is determined through precision-cost elasticity analysis. Finally, a multi-objective optimization function is constructed, integrating navigation distance, fuel cost, and operational time matching. The optimal navigation route is solved using SA-PSO. This invention effectively solves the problems of systematic bias and insufficient resource adaptability in traditional surveys, significantly improves sampling representativeness and navigation operation efficiency, reduces survey costs, and can be dynamically adjusted according to resource conservation and fisheries production needs.

The utility model provides a kind of steel sheet granary, comprising: warehouse body, mounting bracket, sampling mechanism and collection mechanism, wherein, the lower part of warehouse body is provided with support frame;The mounting bracket is connected in the upper portion of warehouse body one side;The sampling mechanism is arranged between the mounting bracket and the warehouse body;The collection mechanism is detachably connected in the lower end of the first sampling pipe.The steel sheet granary of the utility model embodiment, the grain in granary can be stratified sampling by sampling mechanism, and different samples are classified and collected by collection mechanism, so that subsequent staff can accurately detect and analyze grain samples, intuitively obtain the moisture mildew and insect damage of grain at different depths in warehouse body, and timely corresponding measures are facilitated, which greatly reduces the risk hidden danger.

A multi-stratified sampling system using a machine learning algorithm, according to one embodiment of the present invention, comprises: a sample size calculation unit for calculating a sample size on the basis of the size of a target population; a multi-stratified sampling unit for performing multi-stratified sampling on the target population on the basis of the calculated sample size; and a multi-stratified sample group calculation unit for calculating a multi-stratified sample group from the target population on the basis of the performed multi-stratified sampling.

This invention relates to the field of water pollution detection technology and discloses an integrated device for in-situ detection and stratified sampling of groundwater pollution. The device includes a drill pipe, a stratified sampling mechanism, and an in-situ detection mechanism. It also includes a stratified sampling tube assembly fixedly installed inside the drill pipe. The stratified sampling tube assembly consists of multiple sets of sampling tubes arranged vertically. Each set of sampling tubes has a first mounting plate above it, and each first mounting plate has a second mounting plate above it. A drive conversion mechanism is rotatably mounted on the top of each second mounting plate. The technical solution of this invention, through the configuration of the stratified sampling tube assembly, the drive conversion mechanism, and the mechanical linkage mechanism, achieves automated and integrated operation of stratified sampling by the stratified sampling mechanism corresponding to each sampling tube and in-situ detection by the in-situ detection mechanism. This enables rapid in-situ determination and targeted sampling of groundwater pollution conditions, improves work efficiency, and meets the requirements of rapid on-site investigation.

The application discloses a kind of financial risk analysis method and system based on big data analysis model, belong to financial tax analysis technical field, the financial risk analysis method includes obtaining the multidimensional enterprise internal financial tax data of target enterprise in set time period, and the multi-source external correlation data from industry database, supply chain platform, macroeconomic index library.The application is replaced by adopting the directional stratified sampling based on feature importance instead of complete random screening, the data of input model is fundamentally ensured to be the core risk driving factor of high prediction value, and noise and irrelevant features are effectively filtered.Enhanced financial risk analysis model constructed in combination with multidimensional correlation analysis can more comprehensively and deeply depict the complex mechanism of financial risk formation.The two improvements work together to overcome the problems of large model fluctuations and unstable early warning caused by data blindness and single analysis dimension in the prior art, making the analysis results more reliable and accurate.

This invention discloses a method, device, storage medium, and program product for intra-diurnal precipitation simulation based on a weather generator and precipitation decomposition. It establishes a complete precipitation simulation framework from diurnal to hourly scales, organically combining the weather generator and precipitation decomposition algorithm to overcome the shortcomings of traditional methods where each component is independent. An empirical rank rearrangement method is used for post-processing of the simulation sequence, effectively restoring the temporal correlation structure of precipitation and significantly improving the simulation accuracy of statistical characteristics such as the first-order autocorrelation coefficient. A joint stratified sampling strategy based on precipitation and temperature is employed for historical date matching, fully utilizing the covariant relationships between meteorological elements and improving the representativeness of hourly precipitation distribution patterns. An ensemble simulation framework is used, employing multiple realization and multiple sampling mechanisms to quantify simulation uncertainties, providing probabilistic forecasting capabilities for hydrological model applications and risk assessment.

This utility model relates to the field of water conservancy and hydropower testing technology, specifically a sample collection device with stratified sampling function. It includes a moving component, an adjusting component mounted on one side of the moving component, a winding component mounted on one side of the adjusting component, a cable mounted on the inner wall of the winding component, and a collection component mounted on one end of the cable. The moving component includes a base plate, a support plate and a fixing tube mounted on the top of the base plate, an L-shaped plate slidably mounted on the inner wall of the fixing tube, and a push rod mounted on the rear end of the L-shaped plate. The improved sample collection device, employing the adjusting and winding components, allows for automatic lifting of the collection component during use, enabling quick and convenient sample collection and improving work efficiency. The use of the collection component also allows for stratified sampling, enhancing the device's applicability.

This invention discloses a sampling and monitoring method, system, equipment, and storage medium for geological disaster prevention and control, relating to the field of geological disaster monitoring and prevention technology. The method includes generating a cross-depth adaptive discrimination time window based on cross-correlation extreme values ​​and weighted combinations, and obtaining the time window length; generating event markers within the depth based on hierarchical thresholds within the cross-depth adaptive discrimination time window; obtaining inter-layer scores based on the consistency of change directions between adjacent depths; outputting a set of trigger layers when the scores meet the criteria and the number of trigger layers is greater than or equal to a set value; adding a layer above and below the trigger layer set to form a target depth set; determining the event window based on the maximum value of the adaptive discrimination time window and a proportional coefficient; performing stratified sampling and archiving at the target depth; and using chained hashing to bind lag parameters, inter-layer scores, and window parameters to the monitoring sequence. The method described in this invention achieves accurate identification and reliable evidence storage of geological disaster monitoring events through adaptive time windows and inter-layer consistency discrimination.