21 results about "Support vector regression machine" patented technology

The invention relates to the technical field of integrated learning deformation prediction, in particular to a double-weight stacking deformation prediction integrated model modeling method based on complex sample guidance, and the method comprises the steps: building five heterogeneous base models: CLAnet, RBF, MLP, CNN and XGBoost; establishing a support vector regression machine as a meta-model; constructing a secondary integration model on the basis of a stacking framework; a complex sample-oriented five-fold cross validation strategy is adopted to optimize training set distribution, and learning of complex samples is dynamically enhanced; by calculating an error index of each base learner, an initial weight is manually allocated to the base model before the meta-model automatically and implicitly allocates the weight, and metadata set distribution is optimized; and a whale optimization algorithm is introduced to adjust hyper-parameters of each model. According to the integrated prediction model, the problems that a single model is insufficient in generalization ability to complex samples and limited in modeling ability of a multivariable coupling relation can be solved by integrating different learning modes of each base model to features and by means of the quadratic fitting ability of the meta-model.

The invention discloses a decoupling method of an integrated six-dimensional force sensor for a combine harvester, and belongs to the technical field of sensor application. The sensor adopts a Y-shaped three-beam radial symmetric structure, and highly integrates a belt wheel function and a six-dimensional force detection function: establishing an elastomer structure simplified mechanical model, establishing a semi-analytical method model, and analyzing to obtain a strain distribution rule under a uniaxial load. Acquiring a sensor geometric structure and a strain gauge measurement position which meet design requirements according to a strain distribution rule; a multi-output least square support vector regression machine is adopted, hyper-parameters are optimized through grid search, a high-precision mapping model from an electric signal to a six-dimensional load is constructed, the decoupling precision is greatly improved, the whole process innovation from the structure to stress analysis to decoupling is realized, and the method is suitable for large-scale popularization and application. The device is especially suitable for agricultural machinery scenes with limited space and complex load, such as combine harvesters.

The invention provides a cosmetic formula development method, system and device and a storage medium, and the method comprises the steps: obtaining a target demand, and screening out key components meeting the target demand based on a pre-constructed cosmetic knowledge graph; constructing a mathematical model for describing the relationship between the component proportion of the key component and the target performance index by using a response surface method, and carrying out synergistic / antagonistic effect analysis based on the mathematical model to determine a combination formula with a synergistic effect; the combination formula comprises the combination of key components and the component concentration; and performing a uniform test on the combined formula to obtain uniform test data, taking the uniform test data as input of an efficacy prediction model based on a support vector regression machine, and outputting an optimal target formula based on the efficacy prediction model. According to the method, the design and development efficiency of new cosmetic products can be greatly improved, and particularly, the effect and quality of cosmetics are strictly and accurately controlled.

The invention belongs to the technical field of Bluetooth positioning, and particularly relates to a Bluetooth indoor propagation model correction method of a support vector regression machine, comprising the following steps: S1, determining a Bluetooth indoor initial propagation model to be corrected, the initial propagation model being an indoor propagation model improved based on a free space propagation model or a logarithmic distance path loss model, input parameters of the initial propagation model at least comprise a linear distance between a Bluetooth transmitting end and a Bluetooth receiving end and the number of walls on a propagation path, and output parameters are propagation loss values of Bluetooth signals; according to the support vector regression (SVR) Bluetooth indoor propagation model correction method provided by the invention, the problems of single feature dimension, unquantized time sequence fluctuation, poor regional adaptability and the like in the prior art are effectively solved by innovatively fusing multi-modal data collaborative acquisition, time sequence statistical feature modeling and partition adaptive correction strategies.

The application discloses a kind of evaluation method and device of cable thermal fault, its method includes: obtaining the basic data and historical temperature measurement data of to-be-measured cable, according to the preset environmental factor structure condition and preset boundary condition, the 3D model of cable corresponding to basic data is constructed, based on historical temperature measurement data, whether the error of 3D model and actual cable is greater than preset threshold value is judged, if yes, then adjust boundary condition, return to execute the model construction step, if no, then the attribute data of 3D model is calculated by orthogonal control method, based on attribute data, in combination with preset support vector regression machine clustering model, the evaluation result data of to-be-measured cable thermal fault is obtained.There is advantage in solving the technical problem that existing cable thermal fault monitoring method cannot reflect the change situation of contact resistance, and the evaluation efficiency of cable thermal fault is improved.

The invention discloses a shield cutter friction coefficient real-time monitoring and wear trend prediction method and device, and the method comprises the steps: transmitting a high-frequency pulse wave through an ultrasonic sensor embedded in a cutter working surface, collecting a reflection signal of a cutter-rock soil interface, extracting the multi-dimensional acoustic characteristics of the reflection signal, and synchronously collecting the temperature of a friction region; compensating the multi-dimensional acoustic features through a nonlinear regression model, inputting the compensated features into a support vector regression machine, outputting a real-time friction coefficient, and predicting a wear trend through an autoregressive integral moving average model based on a historical sequence of mu; the device comprises a sound-temperature integrated probe, an embedded processing unit and an anti-vibration sealing structure, and provides core data support for shield cutter service life management, maintenance strategy optimization and digital construction.

The present invention discloses a stealth target detection method based on an improved fuzzy support vector regression machine. During the process of detecting hidden targets with a high-sensitivity millimeter-wave radiometer, the radiometer will be affected by background noise, resulting in erroneous detection of stealth targets. The present invention incorporates the suppression of outliers and noise, and simultaneously adopts particle swarm optimization to fuse the penalty parameters and kernel parameters in Levenberg-Marquardt's fuzzy support vector regression model. The fuzzy support vector regression machine model is used to invert the measurement parameters of stealth targets in ground objects detected by the millimeter-wave radiometer. The target's emissivity is calculated based on the inverted target brightness temperature and ambient temperature, thereby realizing the detection of stealth targets in ground objects. This method reduces the influence of noise and outliers on target detection results by introducing a fuzzy membership function into the support vector regression machine, and optimizes the parameters of the fuzzy support vector regression machine using an optimization algorithm, thereby improving the accuracy of stealth target detection in ground objects.

The invention provides a scrap steel quality prediction method and system and a server, and relates to the technical field of metallurgy, and the method fully utilizes heat balance and material balance in the smelting process of a steel furnace, and constructs a scrap steel prediction empirical model on the basis of a scrap steel metallurgy mechanism model. And the twin support vector regression machine and the whale swarm algorithm are combined with the production data to realize accurate prediction of the quality of the waste steel required by the next heat, so that the production efficiency can be greatly improved, and the workload of operators is reduced.

The application provides a multi-attribute target decision evaluation method for enhanced corrosion resistance design of a bent bolt, which is based on a multi-attribute decision analysis theory, and a support vector regression (SVR) machine learning algorithm is innovatively introduced to solve weight coefficients of various performance evaluation indexes, so as to obtain an objective and accurate comprehensive performance evaluation result. Compared with a traditional subjective weighting method, the biggest advantage of the method is that the method can fully explore index correlation rules and weight distribution characteristics contained in historical evaluation data, and the method realizes adaptive solving from data to weights through the machine learning algorithm, and the method has the characteristics of strong objectivity, good adaptability and strong interpretability. The method can also be widely applied to corrosion resistance performance evaluation of a new coating system, and has important guiding significance for optimization and iteration of a coating formula and process parameters.

The invention discloses an intelligent peak flow rate early warning method and a related device. The method comprises the following steps: measuring a peak expiration flow rate of a subject in an expiration process by using a peak flow rate measuring device; calculating the real-time flow of the pressure data by using a Bernoulli equation; the maximum value in the real-time flow is determined to serve as the peak expiration flow rate, and the time point corresponding to the peak expiration flow rate is recorded; inputting the peak expiratory flow velocity, the time point, the personal information of the subject and the medication record of the subject into a trained weighted support vector regression machine model, and predicting the predicted peak expiratory flow velocity of the subject in the future preset time; and generating an early warning signal when the minimum value or the decrease amplitude of the predicted peak expiratory flow rate exceeds a set threshold value. According to the method, the predicted peak expiratory flow rate of the subject in the future preset time is predicted according to the peak expiratory flow rate, the time point, the personal information of the subject and the medication record of the subject, the patient compliance is remarkably improved, and the early warning scheme has higher practical value.

This invention relates to the field of vancomycin pharmacokinetic technology, specifically to the design of a vancomycin clearance prediction scheme based on a VPB ensemble model. This method, based on an ensemble learning strategy, constructs a blending ensemble model to predict vancomycin clearance in adult Chinese patients. First, a variational autoencoder is used to amplify the original sample data to increase data diversity. Then, a particle swarm optimization algorithm is introduced to optimize the parameters of multiple base learners, and the prediction results of the optimized base learners are used as new feature inputs. Finally, a support vector regression machine is used as a meta-learner to integrate and model the above features, forming the final vancomycin clearance prediction model. The VPB model constructed in this invention achieves a determination coefficient R² exceeding 0.9 on both the test and training sets, demonstrating superior prediction accuracy compared to population pharmacokinetic models.

The invention relates to a flexible machine finger instrument control method and system based on visual guidance, and relates to the technical field of robot control. Performing catheter tail end feature point segmentation of the lightweight convolutional neural network, and calling a PnP algorithm to solve a rotation variable and a translation variable under a camera coordinate system; three-dimensional coordinate system positioning information is called to serve as target track input, and contact type pressure sensing waveform data of the flexible machine finger and the catheter wall are output; synchronously loading original waveform data of a pressure sensor in the conduit, executing wavelet packet decomposition to extract frequency band energy distribution of dual-channel signals, and inputting detail coefficient components into a support vector regression machine for dimensionality reduction so as to obtain viscosity change rate and flow resistance gradient data of fluid in the conduit; and outputting a dredging control sequence of the vibration frequency, the amplitude and the action duration of the flexible machine finger. According to the invention, cooperative control of intelligent pre-judgment of the blockage risk and active dredging of the flexible machine finger is realized.

The invention discloses a nondestructive testing method and system for internal cracks of a prebaked raw anode block, and mainly relates to the technical field of industrial nondestructive testing. Sound vibration is generated through force hammer excitation, a block sound vibration response signal is collected, a frequency spectrum is obtained through band-pass filtering and fast Fourier transform, a plurality of main peak frequencies and amplitudes are extracted as initial features, and dimensionality reduction is carried out through principal component analysis. And adjusting parameters of a support vector regression machine by adopting a particle swarm optimization algorithm, and establishing a mapping relationship between the dimension reduction characteristics and the crack area and position height, so as to realize quantitative prediction of the crack parameters. The method has the beneficial effects that the detection speed is high, the precision is high, the positioning and quantification integrated analysis of cracks can be realized, the method is suitable for the automatic quality detection of the prebaked anode green block, and the qualified rate of final products is improved while the comprehensive energy consumption of production is reduced.

The invention relates to a river reservoir suspended sediment remote sensing recognition system and method based on water color difference, and belongs to the technical field of water ecological environment monitoring. The method comprises the following steps: firstly, acquiring a multispectral satellite image through a GEE cloud platform, and extracting reflectivity data of main and branch flows through preprocessing such as atmospheric correction and image fusion; performing space-time registration on the in-situ turbidity measured value and the satellite reflectivity to construct a training data set; aiming at different water color characteristics of the main stream and the branch stream, respectively establishing a support vector regression machine learning model of the main stream and a multi-hydrological-period empirical model of the branch stream, and realizing high-precision zoning inversion of turbidity; and finally, identifying the suspended sediment water body through turbidity grading, and generating a spatial-temporal distribution map. The method effectively solves the problem of water color difference of the main stream and the branch stream caused by reservoir regulation, remarkably improves the precision and the reliability of remote sensing monitoring of the suspended sediment, and provides technical support for river reservoir sediment treatment and water environment management.

The application discloses a decoupling method of an integrated six-dimensional force sensor for a combine harvester and belongs to the technical field of sensor application. The sensor adopts a "Y" type three-beam radial symmetric structure, highly integrates a belt wheel function and a six-dimensional force detection function, establishes an elastic body structure simplified mechanics model, establishes a semi-analytical method model, analyzes and obtains strain distribution laws of X, Y and Z under single-axis load, and obtains a sensor geometric structure and a strain gauge measurement position meeting design requirements according to the strain distribution laws; a multi-output least square support vector regression machine is adopted, hyperparameters are optimized through a grid search, a high-precision mapping model of an electric signal to six-dimensional load is constructed, the decoupling precision is greatly improved, a whole-process innovation from structure to stress analysis and then to decoupling is realized, and the method is especially suitable for agricultural mechanical scenes such as a combine harvester and the like which are space-limited and have complex load.

The application discloses a combined estimation method for tire vertical force and side force based on in-utero strain analysis, and comprises the following steps: S1, establishing a tire finite element three-dimensional model; S2, calculating a tire grounding angle φ and a grounding length L based on the tire finite element three-dimensional model established in the step S1 according to in-utero strain; S3, estimating a tire vertical force F based on the tire grounding angle φ and the grounding length L calculated in the step S2 by using a support vector regression machine; S4, simulating and calculating a tire side deflection working condition by using the tire finite element three-dimensional model, taking a last valley value h2 of a side deflection circumferential strain difference curve as a characteristic of a side deflection force F ; S5, establishing a combined estimation model for the vertical force and the side deflection force based on the characteristic h2 of the tire vertical force F and the side deflection force F, and estimating an actual side deflection force of the tire. c c z y z y The application is applicable to static load, rolling and side deflection working conditions, can accurately estimate the vertical force and the side deflection force, and the error between an estimated value and a finite element simulation value is less than 3%.​​​​​

This invention provides a method, apparatus, equipment, and medium for predicting shale gas content. The shale gas content prediction method includes the following steps: confirming the geological parameters of the target shale and performing grey relational analysis on the geological parameters and shale gas content to identify the main controlling factors of shale gas content; establishing a support vector regression (SVR) model, optimizing the parameters of the SVR model, outputting the optimal parameters, and constructing an optimal SVR model; training the optimal SVR model based on the main controlling factors of shale gas content to obtain the shale gas content prediction model; and predicting the gas content of the target shale based on the shale gas content prediction model. This invention uses grey relational analysis to select the main controlling factors and optimize the support vector regression (SVR) model, thus solving the problem that the performance of the SVR model depends on the selection of its hyperparameters.

The application is a river reservoir suspended sediment remote sensing recognition system and method based on water color difference, belonging to the technical field of aquatic ecological environment monitoring. First, multispectral satellite images are obtained through the GEE cloud platform, and dry and branch stream reflectivity data are extracted after preprocessing such as atmospheric correction and image fusion. Then, in-situ turbidity measurement values are spatiotemporally matched with satellite reflectivity to construct a training data set. According to the different water color characteristics of the dry and branch streams, a support vector regression machine learning model for the dry stream and a multi-hydrological period empirical model for the branch stream are respectively established to realize high-precision partition inversion of turbidity. Finally, suspended sediment water bodies are identified through turbidity grading, and a spatiotemporal distribution atlas is generated. The application effectively solves the problem of water color difference between dry and branch streams caused by reservoir regulation, significantly improves the precision and reliability of suspended sediment remote sensing monitoring, and provides technical support for river reservoir sediment control and water environment management.

The application provides a scrap steel quality prediction method, system and server, relates to the metallurgical technology field, and fully utilizes heat balance and material balance in a smelting process of a steelmaking furnace. An experience model of scrap steel prediction is constructed on the basis of a scrap steel metallurgical mechanism model. Twin support vector regression machines and whale swarm algorithms are combined with production data to realize accurate prediction of required scrap steel quality of a next furnace, so that production efficiency can be greatly improved, and the workload of an operator can be reduced.

The invention relates to the technical field of oil sand and super heavy oil reservoir development, in particular to a variable well position deployment SAGD optimization method and device. According to the method, the variable well position deployment SAGD development mode different from a conventional SAGD development mode is considered, namely, the SAGD encrypted production horizontal well development mode and the dislocation well pair SAGD development mode are considered, and the SAGD development effect of the oil sand and super heavy oil reservoir can be further improved. By utilizing secondary coupling of a support vector regression method and a genetic algorithm in the field of machine learning, a variable well location deployment SAGD optimization design workflow based on machine learning is established, and rapid, reliable and intelligent optimization design of related well location deployment parameters and operation parameters of a variable well location deployment SAGD development mode can be realized. The method is suitable for well position deployment and operation parameter design of an oil sand and super heavy oil reservoir variable well position deployment SAGD development mode, and technical support is provided for efficient decision making and benefit development of a production site.

The application discloses a new drilling potential evaluation method based on a mixed feature selection, and is characterized in that the number of newly produced wells in a historical interval is calculated, and reservoir development history data is marked; the correlation influence of each feature in the reservoir development history data is analyzed by using a Pearson product-moment correlation coefficient, individual fitness values are calculated, an elite selection mechanism is adopted to add individuals with larger fitness values and individuals generated after random selection and crossover to a sub-population, the population number is kept stable, and iteration is performed until an optimal feature combination of the oilfield history data is obtained; a support vector regression machine is used to mine deep information of the oilfield history data, and new drilling potential evaluation is realized. The application calculates the correlation influence of features by using the Pearson product-moment correlation coefficient, guides the selection of the optimal feature combination of the oilfield history data, speeds up the execution speed of the feature selection, adds the elite selection mechanism, increases the search precision of the feature combination of the oilfield history data, and improves the accuracy of the new drilling potential evaluation.