27 results about "Multivariable linear regression" patented technology

The application discloses a general optimization method for 3D printing sugar-reduced food, comprising the following steps: S1. Defining and measuring core parameters: including porosity P of a 3D printing food matrix, odor concentration C of an odor active component accounting for a total mass proportion of the food matrix, sucrose reduction amount Suc compared with a traditional full-sugar food, and a sensory evaluation index S of a sensory evaluator on a sweetness similarity degree between the sugar-reduced food and the traditional full-sugar food; S2. Constructing a quantitative model: under the condition of setting the sucrose reduction amount Suc value in advance, a plurality of different P, C and S corresponding data are obtained through experimental design, and based on the data, a quantitative model of the sweetness similarity S about the porosity P and the odor concentration C is obtained by using a multiple linear regression method; and the problem that the prior art cannot clearly guide how to set the printing parameters and the formula to obtain the best comprehensive sensory experience under the premise of ensuring a certain sugar reduction ratio is solved.

This invention discloses a method for optimizing flue gas NO based on thermal parameters. x The emission control method includes the following steps: online measurement of thermal parameters for each stage of pellet roasting; analysis of the thermal parameters for each stage and their relationship with the NO inlet of the SCR denitrification system. x The correlation between concentrations was determined to establish the relationship between NO at the SCR denitrification inlet and each process stage. x Significantly correlated thermal parameters with concentration; constructing and solving for NO at the SCR denitrification inlet. x A multiple linear regression mathematical model for NO concentration was used to predict the NO concentration at the SCR denitrification inlet. x The trend of NO concentration change; identify the relationship with the SCR denitrification inlet NO. x The single factor most relevant to the concentration was established, and the relationship between the single factor and the NO inlet of the SCR denitrification system was established. x A linear relationship between concentrations was established to adjust NO x A univariate linear regression mathematical model for NO concentration is used to regulate NO. x Concentration. This invention can predict NO concentration at the SCR denitrification inlet. x The concentration trend can be observed, and the NO inlet of the SCR denitrification system can be precisely controlled. x Concentration, effectively achieving NO concentration in pellet roasting x Emissions are controlled throughout the process to achieve the goals of energy conservation and environmental protection.

The method, system, and equipment for screening and optimizing reservoir gas injection parameters based on response surface methodology include: determining the gas injection parameters affecting reservoir gas drive recovery and assigning initial weights; selecting basic spatial sample points using the central composite method; establishing a response surface function; constructing a response surface function model; validating the response surface model and verifying its accuracy; obtaining the gas injection parameter variable set; obtaining the distribution results of the gas injection parameter variable set converted into weight coefficients; and obtaining the quadratic response distribution of the weight coefficients through quadratic response surface analysis to obtain the discrimination criteria for gas drive in fractured-vuggy carbonate reservoirs. The reservoir gas injection parameter screening method provided by this invention uses response surface analysis combined with multiple linear regression to screen reservoir gas injection parameters. It uses graphical technology to display the functional relationships, providing intuitive graphs to clearly identify the optimization region, allowing researchers to observe and select the optimal conditions in experimental design.

The present application provides a pore pressure prediction method based on rock physics modeling and multiple linear regression, relates to the oil and gas exploration and development technical field, and the method comprises the following steps: S1: selecting a plurality of reference wells in a secondary structural unit for overpressure analysis; S2: pre-processing the logging data of the reference wells and analyzing the overpressure causes; S3: performing fluid replacement by using the Gassmann equation, and performing solid replacement by using the Brown-Korringa theory, and calculating the rock elastic modulus; S4: selecting an anisotropic soft pore model to calculate the rock effective velocity; S5: performing sensitivity analysis on the elastic parameters and the pressure coefficient; S6: constructing a multiple linear regression model with the elastic parameters having the best correlation with the pressure coefficient, and predicting the pore pressure; and S7: comparing and verifying the prediction result with the Eaton method. The present application avoids the problem of errors caused by relying on the normal compaction trend line, fits the elastic parameters having good correlation with the pressure by using multiple linear regression, comprehensively considers the influence of multiple variables, and has higher prediction and interpretation capability.

This invention discloses a method, system, equipment, and medium for quantitative research on coal quality robustness coefficient, relating to the field of coal mine safety engineering technology. The method includes: collecting blocky coal samples of different ranks, pre-processing them, and measuring the robustness coefficient using the drop hammer crushing method; obtaining carbon structure-related data through X-ray photoelectron spectroscopy and solid-state nuclear magnetic resonance testing, combined with software analysis; performing linear fitting analysis on the robustness coefficient and carbon structure-related data, calculating the influence weights, and constructing a multiple linear regression equation to achieve quantitative research on the coal quality robustness coefficient. This invention establishes a relationship equation between the robustness coefficient and carbon skeleton structure characteristics based on the microscopic carbon skeleton structural features and robustness coefficient testing and analysis methods of coal, providing a microscopic perspective for the study of coal and gas outburst mechanisms and having practical guiding significance for safe mine production.

The present application relates to the technical field of combustion vibration prediction, and more particularly to an agent-driven multiple linear regression prediction and residual abnormal parameter alarm method, which proposes the following scheme: through the agent, real-time operation data is input into a pre-trained linear regression model to generate a combustion vibration prediction value, and through the calculation of the residual between the actual and predicted vibration values, the operation condition of the gas turbine is analyzed. Through the data processing classification layer, relevant parameters are extracted, the non-linear characteristic linearization is realized by combining the physical model fine-tuning kernel function, and the regression model performance is optimized. In addition, the residual analysis and long-term change trend are used to identify abnormal patterns, so as to realize the accurate monitoring and abnormal alarm of the combustion vibration. The method improves the prediction accuracy and calculation efficiency, and guarantees the stable operation of the gas turbine.

This invention relates to a method for estimating surface dead combustible load in subtropical forests using multi-source remote sensing and machine learning. The method includes: acquiring and preprocessing multi-source remote sensing data and auxiliary data within a target area; acquiring preprocessed multi-source remote sensing data and auxiliary data; extracting features from the preprocessed multi-source remote sensing data and auxiliary data to obtain feature variables; selecting variables based on the feature variables and measured surface dead combustible load components; constructing a combustible load prediction model using multiple linear regression and machine learning models; evaluating the combustible load prediction model to obtain the optimal model; and using the optimal model to perform regional inversion of surface dead combustible load for all vegetation types within the target area to obtain spatial distribution data of surface dead combustible load. This invention combines multi-source remote sensing with machine learning to provide methodological support for fire risk assessment and precise combustible management in subtropical forests.

The application provides a kind of multi-model fusion inverse calculation pumping unit well dynamic liquid level method, system, electronic equipment and storage medium, the method comprises: based on oilfield data lake and oil and gas production internet of things database, obtains dynamic liquid level inverse calculation parameter data and is stored in dynamic liquid level inverse calculation system standard library;With the data in the dynamic liquid level inverse calculation system standard library, respectively through mechanism analysis method model, AdaBoost algorithm and MLP&RNN algorithm, inverse calculation obtains dynamic liquid level depth H1, H2 and H3;Adopt multivariate linear regression method, combine the three inverse calculation results, construct dynamic liquid level fusion refined calculation model, to realize the accurate prediction of actual dynamic liquid level depth H.The application can accurately obtain the dynamic liquid level of pumping unit well, improve the accuracy and adaptability of dynamic liquid level parameter, provide accurate data basis for dynamic analysis to take measures and determine reasonable working system.

The application discloses a cognitive impairment development trajectory structural magnetic resonance classification method based on multiple space scales, and comprises the following steps: selecting a first-level brain atlas, splitting and merging the first-level brain atlas into second and third-level atlases according to spatial structure relationships of sub-brain regions; pre-processing a structural magnetic resonance image; extracting seven structural magnetic resonance features of each sub-brain region in the three levels; using the three-level atlases, constructing a three-space-scale intracerebral layer connection network through multiple linear regression, combining the three-space-scale features, and then considering the relationship among the three levels to construct an interlayer connection network with a space scale of three; selecting a connection with a difference from the obtained connection matrix, recursively eliminating the features to obtain prediction features, taking single-space-scale features, three-space-scale fused features and interlayer features of multiple space scales as inputs, respectively, and using a classifier to evaluate the feature performance; and finally, the optimal classification features obtained have a further understanding of the widely recognized brain region connection.

The application relates to a meat quality evaluation method based on single-factor variance analysis, which comprises the following steps: acquiring multi-sample data, performing standardization processing on the multi-sample data, sequentially performing oneway-ANOVA analysis and principal component analysis on the multi-sample data after the standardization processing, and generating the weight of each index in target classification; extracting the first preset number of indexes in each target classification for one-sided evaluation, obtaining individual single-item scores, taking the individual single-item scores as dependent variables to construct a multiple linear regression equation model, acquiring individual single-item score prediction values, constructing a comprehensive multiple linear regression equation model, and obtaining individual comprehensive scores. The application can quickly, accurately, conveniently and objectively evaluate pork quality.

The application discloses a microstrip line passive intermodulation prediction method based on a nonlinear coefficient R n , and belongs to the technical field of passive intermodulation prediction. The application obtains a parameter nonlinear coefficient R2 based on an edge defect index of a microstrip line, applies a multiple linear regression method, and establishes a passive intermodulation prediction model to predict third-order passive intermodulation power levels of the microstrip line under different processes. The mapping relation between the microscopic defect index and the nonlinear coefficient is substituted into an expression for solving third-order passive intermodulation transmission and reflection power, the error between the prediction and the test result is within the range of ±6 dB, and a feasible method is provided for the prediction and analysis of the third-order passive intermodulation power of the microstrip line.

This invention relates to the field of collaborative transportation using public transport and drones, and discloses an optimized method for express delivery allocation in urban public transport and drone collaborative transportation. The steps are as follows: collecting urban traffic flow information data; extracting clustered bus trajectory data features, constructing a multiple linear regression equation to predict road segment travel time, and continuously optimizing the equation based on actual results to reduce prediction errors; obtaining the operating time of public transport in each city, the Euclidean distance from each POI point to the integrated transfer point, and constructing a two-layer optimization framework using a three-way optimization algorithm on the outer layer and a genetic algorithm on the inner layer to realize the express delivery allocation scheme; loading express deliveries according to the allocation scheme for each batch, and determining whether to perform transfer based on real-time traffic data. This further improves the overall efficiency and intelligence level of urban last-mile logistics.

The application discloses a kind of CO2 drive oil reservoir oil production well wellbore phase transition point calculation methods, comprising the following steps: S1: calculating the bubble point temperature-pressure relationship curve and wax precipitation temperature-pressure relationship curve of target well;S2: calculating the temperature-depth relationship curve and pressure-depth relationship curve of target well under different working conditions;S3: calculating the bubble point temperature-depth relationship curve and wax precipitation temperature-depth relationship curve under different working conditions;S4: calculating the gas-liquid phase transition point depth and wax precipitation phase transition point depth under different working conditions;S5: by multiple linear regression, obtain calculation formula and calculation formula;S6: the production condition of target well measured is substituted into and calculation formula, the phase transition point of target well under this production condition can be calculated.This method realizes the fast and accurate prediction of CO2 drive oil production well wellbore phase transition point.

The application discloses a deep coal bed gas reservoir high brittleness sweet spot area prediction method, system and equipment, relates to the oil and gas and coal bed gas geophysical exploration field, and the method comprises the steps of: taking dynamic elastic modulus as the independent variable, taking coal bed fracture pressure as the dependent variable, adopting a multiple linear regression method to perform fitting, determining a partial regression coefficient, and thereby constructing a brittleness index model; based on original data of a target area deep coal bed gas reservoir, adopting prestack simultaneous inversion technology and rock physical formula to perform inversion and calculation, obtaining a three-dimensional dynamic Young's modulus data body, a three-dimensional dynamic Poisson's ratio data body and a three-dimensional dynamic shear modulus data body, and substituting into the brittleness index model to obtain a three-dimensional brittleness index data body; according to a brittleness index threshold, dividing a region corresponding to the three-dimensional brittleness index data body to determine a high brittleness sweet spot area. The application can improve the accuracy of predicting the high brittleness sweet spot area of the deep coal bed gas reservoir.

This application provides a method and electronic equipment for predicting the production of SAGD drainage from a combination of vertical and horizontal wells, belonging to the technical field of enhanced oil recovery in heavy oil reservoirs. The method includes: establishing a numerical model for SAGD drainage from vertical and horizontal wells based on geological condition screening criteria for heavy oil reservoirs and combined with geological parameters of the oilfield block; analyzing the relationship between geological factors, engineering factors, and production based on the SAGD numerical model, and determining the value ranges of geological and engineering factors; determining the main controlling geological factors and main controlling engineering factors based on the impact of geological and engineering factors within different value ranges on production; establishing a multiple linear regression equation between the main controlling geological factors, main controlling engineering factors, and production using multiple linear regression, and using the multiple linear regression equation as the production evaluation model for the SAGD drainage from vertical and horizontal wells; and predicting the production of SAGD drainage from vertical and horizontal wells based on the production evaluation model. This application improves the accuracy of SAGD drainage production prediction.

The present application relates to the field of fiber-optic sensor fiber-optic gyroscope, and particularly relates to a kind of combination structure sensor and leap time fiber-optic gyroscope temperature compensation method and system, comprising: structural temperature data is obtained by acquisition;Obtain the leap time data of optical wave transmission in optical fiber ring one circle;Obtain the zero bias shift data and scale factor data of fiber-optic gyroscope;Leap time data is converted, and the optical fiber ring optical path temperature data is obtained;Based on structural temperature data, optical fiber ring optical path temperature data, zero bias shift data and scale factor data, multiple linear regression is carried out, and temperature error compensation model is established;Real-time acquisition structural temperature and optical fiber ring optical path temperature are input into the temperature error compensation model, and zero bias shift prediction value and scale factor prediction value are obtained, and the original angular rate is compensated using zero bias shift prediction value and scale factor prediction value, and compensation angular rate is obtained.The present application can improve the performance of fiber-optic gyroscope temperature compensation.

The application discloses a quick quantitative method for sulfates in saline soil based on mid-infrared spectrum combined with multivariate linear regression, and belongs to the technical field of saline soil detection, and comprises the following steps: S1, saline soil sample collection and pretreatment; S2, mid-infrared spectrum data collection; S3, joint pretreatment of spectrum data; S4, screening of sulfate-sensitive characteristic wave bands; S5, construction of a multivariate linear regression quantitative model; and S6, quick quantitative detection of a to-be-detected sample.The application adopts mid-infrared spectrum to quantitatively detect sulfates, utilizes the fundamental frequency characteristic absorption peak of sulfate ions in the mid-infrared wave band, and compared with the general frequency absorption of near-infrared, the characteristic peak is more specific, the absorption intensity is higher, and the anti-interference ability is stronger, so that the interference of soil organic matter, clay minerals and other salt can be effectively resisted, and the detection precision is greatly improved.

The application discloses a kind of regional scale whole section soil profile salt's satellite remote sensing partition modeling method, the method comprises: obtaining the vegetation index and soil salt index of remote sensing image;According to the soil profile of different depth, the partition factor of partition is determined, and the vegetation index and soil salt index of the remote sensing image are classified to corresponding partition according to partition factor;The vegetation index and soil salt index of remote sensing image in each partition are carried out multiple linear regression to obtain segmented linear function as final model.The application can obtain soil profile salinization information with high precision, provide advanced technical means for development and utilization of saline soil resources, and has positive significance under the background of global population rapid growth.

This invention discloses an on-orbit method and system for determining the spectral response function of a hyperspectral remote sensor, based on the solar reflection band. The method includes: step S1, acquiring multiple sets of input light sources to obtain a system of multiple linear regression equations; and step S2, calculating the system of multiple linear regression equations using the least squares method to obtain the pixel-by-pixel spectral response function of the remote sensor. This on-orbit method for determining the spectral response function of a hyperspectral remote sensor solves the problem of whether or not on-orbit observation of the spectral response function of a hyperspectral remote sensor is possible. It also shortens the observation cycle from months or even years to minutes, thereby improving the accuracy of satellite-based monitoring of global greenhouse gases, environmental pollutants, aerosols, and other atmospheric components, contributing to the high-quality development of national undertakings such as climate research, environmental protection, and low-carbon emission reduction.

The application discloses a kind of based on RSM's silicon carbide phase transition filling body performance prediction method, it is related to mine filling material technical field, phase change material is phase change microcapsule, include: selecting the response of the performance of filling body and the factor of influencing the performance of filling body;The correlation between the response of the performance of filling body and the interaction between each factor, the correlation between the response of the performance of filling body and the correlation between the response of the performance of filling body are carried out test design and data analysis;According to test data, variance analysis is carried out, and the test results of the response of each corresponding performance are carried out multiple linear regression and binomial fitting analysis using multivariate quadratic polynomial model, verify the regression model of each response variable and the significance of factor;The performance of multiple response variables is comprehensively evaluated using comprehensive desirability, to predict the preferred factor combination.The application can predict the performance of filling body under different proportions in actual engineering based on RSM, can solve the problem of heat damage in deep mine and reduce mining cost during the operation of mine.

The present application relates to the technical field of oil and gas drilling engineering, and specifically discloses a multi-field coupling complex fractured formation fracture width distribution inversion method, based on the theory of porous elasticity, a drilling fluid loss model considering fluid-solid coupling effect is constructed, a roughness correction factor is introduced to correct the traditional cubic law, and the application of Darcy's law in the matrix is improved, a drilling fluid loss mathematical model coupled with the matrix and the fracture is established, through numerical simulation, the influence law of key parameters such as fluid viscosity, pressure difference and fracture width on drilling fluid loss is obtained, and on this basis, a fracture width multivariate linear regression inversion equation with fluid viscosity, pressure difference and cumulative loss as independent variables is established, the inversion equation has high prediction accuracy and practicability, can be used for quickly and accurately inverting the fracture width on site, and provides theoretical and technical support for the understanding of the loss mechanism of multi-scale fractured formation and the optimization of leakage prevention and treatment.

The application relates to a multi-condition frequency coupling impedance identification method and device using multiple linear regression, wherein the method comprises the following steps: obtaining a frequency coupling admittance matrix of a measured device under multiple conditions to obtain a condition parameter variable matrix; based on the condition parameter variable matrix, a permutation importance index of the condition parameter variable is calculated, a first condition parameter variable set satisfying a first preset key condition of the measured device is determined; based on the first condition parameter variable, the multiple collinearity degree of the condition parameter variable is calculated, a second condition parameter variable set satisfying a second preset key condition is obtained, and a multiple linear regression model for identifying the multi-condition frequency coupling impedance of the measured device is constructed. Therefore, the problems in the prior art that when the number of hidden layers of the neural network impedance fitting algorithm exceeds the model generalization requirement, the training set error and the test set error will produce significant deviation, and when extrapolated to unknown conditions, the error will significantly increase and the generalization ability will be insufficient are solved.

The application discloses a Raman selectable pulse fiber laser for acne treatment and relates to the technical field of biomedical engineering treatment, and comprises a hemorrhoid treatment effect monitoring data acquisition module, a hemorrhoid treatment effect monitoring data optimization module, a wavelength precision prediction module, a pulse uniformity prediction module, a laser beam quality evaluation module, a fiber transmission detection module, a hemorrhoid treatment effect evaluation module, a hemorrhoid treatment effect correction module and a Raman selectable pulse fiber laser optimization module. The Raman selectable pulse fiber laser is combined with a random forest algorithm, a neural network algorithm, a multiple linear regression algorithm and a convolutional neural network algorithm to evaluate and calibrate the hemorrhoid treatment effect. The data entry technology, the random forest algorithm, the neural network algorithm, the multiple linear regression algorithm, the convolutional neural network algorithm and the hemorrhoid treatment effect evaluation and calibration technology in the application are closely combined with modern information technology, and the precision of the hemorrhoid treatment effect based on the Raman selectable pulse fiber laser is improved.

This invention discloses a method, system, electronic device, and storage medium for predicting the rehydration ratio. The method includes: collecting process parameters and actual values ​​of the rehydration ratio during the vacuum rehydration process; selecting input variables for a multiple linear regression prediction model of the rehydration ratio based on the process parameters; establishing the multiple linear regression prediction model of the rehydration ratio based on the input variables; fitting the multiple linear regression prediction model with historical data of the input variables and the actual values ​​of the rehydration ratio; collecting real-time values ​​of the input variables and inputting them into the fitted multiple linear regression prediction model to predict the rehydration ratio. This invention has clear quantitative relationships, higher model interpretability, is more intuitive and closer to the principle of rehydration, and, given a fixed leaf blend formula and tobacco pack grade, already possesses the necessary data for calculation, allowing for the calculation of the water addition ratio for different grades of tobacco packs before production.

This disclosure proposes a method and apparatus for predicting the clearing results of inter-provincial spot markets based on ensemble learning, relating to the fields of artificial intelligence and power control technology. The method includes: acquiring meteorological and renewable energy power forecast information for two trading regions; using Pearson correlation coefficients to screen effective data strongly correlated with the clearing results and determining the weights of each effective data point; inputting the weighted effective data in parallel into pre-trained random forest, XGBoost, and LightGBM prediction models to obtain preliminary prediction results from each model; and fusing the preliminary prediction results through a multiple linear regression ensemble learning model to output the predicted clearing results. This disclosure, through multi-model heterogeneous ensemble and data-driven weighting, effectively integrates multi-regional and multi-dimensional features, adapts to complex coupling relationships, and significantly improves the accuracy and robustness of clearing result prediction in complex market environments, providing reliable decision support for power trading and dispatch.

This invention discloses an electric vehicle charging scheduling method based on genetic algorithms and multi-objective optimization to solve the comprehensive optimization problem of charging pile selection and power allocation during long-distance travel. The invention constructs a combined weighted TOPSIS model, which integrates objective entropy weights with user preferences based on multiple linear regression and Fisher classification, obtaining combined weights through the minimum entropy principle to calculate the charging pile suitability score. An improved two-stage genetic algorithm, TPGA, is employed to decompose the optimization process into a charging pile combination selection stage and a dynamic charging capacity allocation stage, achieving efficient path planning and resource allocation under constraints of power continuity, battery capacity, and market balance. Experiments show that compared to existing solutions, this invention achieves significant improvements in core indicators such as convergence speed, additional driving distance, waiting time, and total cost, comprehensively improving user satisfaction and charging decision-making efficiency.

This invention discloses a method for optimizing the heat treatment process of CMT arc cladding of 9Cr1Mo heat-resistant coating on Q235 steel surface. The specific steps are as follows: 1. A large cladding sample is obtained through experiments using the controlled variable method, followed by testing to obtain data on the sample's shear strength and impact absorption energy; 2. The data is preprocessed, a scatter plot is drawn, and a mathematical model of shear strength and impact absorption energy with respect to heat treatment temperature and time is obtained using a multiple linear regression fitting method; 3. The two mathematical models are integrated to obtain a mathematical model for evaluating the comprehensive mechanical properties of the sample. With added constraints, the model is solved, and the optimal combination of heat treatment temperature and time is used to obtain the best comprehensive mechanical properties. This invention ensures that the cladding sample has sufficient interfacial bonding strength and toughness, avoiding deformation, cracking, and premature failure, while also reducing production costs.