66 results about "Principal component regression" patented technology

The invention discloses a method for determining a radix notoginseng extract and contents of five types of ginsenosides in a preparation of the radix notoginseng extract by a Fourier transform near-infrared spectrograph. The method comprises the steps that a sample is detected by an ultra-high performance liquid chromatography, moving phase builds acetonitrile and water gradient elution parameters with separation degrees higher than 1.7 in chromatographic peaks of Rg1 and Re, and the spectrum preprocessing adopts the combination of 2 to 3 methods in Savitzky-Golay polynomial smoothing, second-order differential, Noriis derivative filtering and data normalization; and three preferred wave bands of the five types of ginsenosides are modeled, and a calibration model is built by any one of a partial least squares regression method, a principal component regression method and a multiple linear regression method. The method is used for determining the contents of the five types of ginsenosides in the sample to be detected, the determining result is consistent with a result determined by the ultra-high performance liquid chromatograph basically, the requirements of 'Chinese pharmacopoeia' are met, the accuracy is high, and the determination speed is increased substantially.

The invention discloses a concentration detection method of an organic polymer solution. The method comprises the following steps: obtaining a Raman spectrum of the organic polymer solution with the known concentration; preprocessing a Raman spectrum signal and then obtaining a predication model of the concentration of the organic polymer by a plurality of data processing methods such as a principal component regression method, a partial least square method and an artificial neural network; detecting and preprocessing the Raman spectrum of the organic polymer solution to be detected; and inputting the preprocessed Raman spectrum into the prediction model to calculate the concentration of the organic polymer solution. The method is characterized by being rapid, accurate, safe and environment-friendly, can be applied to on-line and off-line monitoring of the organic polymer and other related products in the production, modification, transportation and recovery processes.

A system and method for monitoring and controlling the properties of fluids is disclosed. The inventive concept employs impedance spectroscopy (IS) measurements, and is suitable for real-time, in situ, monitoring and quality control operations, such as quality control during the manufacture of blended lubricants. IS data are obtained for three or more frequencies, where the lowest frequency is less than 1 Hz and the highest frequency is greater than 1 Hz. These data may be interpreted according to statistical techniques such as Principal Component Regression, analytical techniques such as equivalent circuit modeling, or by a combination thereof. The data analysis provides characteristics, or IS signatures, relating to the properties of the fluid. IS signatures for a test fluid are compared to IS signatures for calibration fluids to determine whether the properties of the test fluid fall within specified limits. Quality control adjustments to the test fluid properties may be performed based on the IS signatures.

The method includes steps: (1) placing fruit sample inside sealed container; after top air portion reaches balance, sampled top air portion is inducted to sensor array reaction chamber; response signal is obtained when reaction takes place between sensor array and top air portion; response signal of sensor is ratio between resistance R after sensor contacts top air and resistance R0 when sensor passes through clean air, i.e. S=R/R0; (2) carrying out detections including compactness, sugar degree, and acidity for detected fruit; (3) using multiple linear regression, major constituent regression, least squares regression, and artificial neural network to build mathematical model of relation between the response signal of sensor and compactness, sugar degree, acidity of fruit sample. The invention extends detection range, lowers interference, increase sensitivity, reliability, and repeatability.

The invention provides a method for distinguishing variety of fritillaria and detecting total alkaloid content of the fritillaria by virtue of near infrared spectrum. The method provided by the invention comprises the following steps: (1) collecting a fritillaria sample; (2) measuring the near infrared diffuse reflection spectrogram of the fritillaria sample, preprocessing the 4000-5000cm<-1> wave band in the spectrogram, and performing cluster analysis on the pre-processed near infrared spectrogram to build a qualitative model; or preprocessing the 4000-7000cm<-1> wave band in the spectrogram, so as to obtain an absorbance, associating the absorbance with the alkaloid content of the sample measured by virtue of bromothymol blue colorimetry, and building a quantitative correction model for detecting alkaloid by one or more methods of a partial least squares method, a principal component regression method and a multiple linear regression method; (3) collecting the near infrared spectrogram of the sample to be measured, after the corresponding preprocessing is performed, distinguishing the variety of the fritillaria and detecting total alkaloid content of the fritillaria by utilizing the built qualitative model or quantitative correction model. The method provided by the invention has the characteristics of fast speed, no damage, environment friendliness and low cost.

The invention discloses a human body intelligent recognition method and a system based on a distributed layout pyroelectric infrared detector. The recognition system consists of layout distributed detectors in a monitoring site and a detector host which transfer information through a wireless communication. The method is to use the detectors to perform a real-time signal process to a detected region; to extract an event window, perform extractions of a time domain feature and a frequency domain feature from signals in the event window, transmit the extracted features to the detector host through the wireless communication, uses the detector host to perform a fusion of a feature-level fusion for information of each detector, and determines whether or not an identifying object is a human body through an algorithm of principal component regression. The human body intelligent recognition method is combined with an information fusion technology of distributed layout detectors and a pattern recognition technology, and can effectively reduce effects of external environment and other creatures on a monitoring system and reduce the probability of misrecognition for a passive pyroelectric infrared detector.

The invention discloses a soft sensing method based on a probabilistic principal component regression model for online detection of butane content of a debutanizer. Principal component regression model is a common modeling approach for soft sensing. However, as noise information of process data is not taken into consideration well, the principal component regression model fails in many real processes. The method introduces a probabilistic modeling approach which converts the conventional principal component regression model into a probabilistic form, which is a soft-sensing model based on probabilistic principal component regression. Compared to the conventional principal component regression model, the probabilistic principal component regression model allows modeling on the process data and the noise information at the same time and provides more complete data information, so that the soft sensing results are more reliable.

The invention discloses a roller kiln temperature prediction integrated modeling method capable of combining a mechanism with data. The method comprises the following steps that: through the analysis of factors which affect temperature change, from a perspective of the temperature change and energy, establishing a mechanism model; then, considering situations that roller kiln sintering is a very complex process, a whole sintering process can not be described through a single mechanism model and the mechanism model has model errors through simplification, establishing a data model to predict model errors so as to make up mechanism output, i.e., utilizing errors to serve as a training sample to establish an error prediction model of a nonlinear time-varying process based on local weighted kernel principal component regression; and finally, combining the mechanism model with the data model to establish a roller kiln temperature prediction integration model. By use of the model, the state change of a process can be better tracked, and a good guidance function is provided for roller kiln temperature control so as to improve the production quality and the percent of pass of a product.

The invention relates to the technical field of water consumption prediction, and specifically relates to a water consumption prediction method and device based on big data, and the method comprises the steps: inputting collected data, and carrying out the data preprocessing through employing a principal component analysis method and a Lasso feature selection model; an annual water consumption prediction model of each region is established; obtaining a water consumption prediction result according to the preprocessed data. According to the invention, a principal component regression model is adopted; the Lasso regression model and the support vector machine regression prediction model are combined. The principal components are further screened by using a lasso algorithm after being analyzed by using the principal components. Compared with the prior art, the method has the advantages that the water consumption with high accuracy can be obtained only by collecting a small number of influence factor data sets, the prediction accuracy of the water consumption is greatly improved, and meanwhile, the water consumption is predicted within a shorter time.

The invention discloses a lithium cobaltate batching system based production state prediction method. The method includes building a health status evaluation index of a lithium cobaltate batching system; and performing prediction on the health index of the batching system by using a double weighted based kernel principal component regression least squares support vector machine algorithm, whereinthe prediction step includes the following steps: firstly, extracting non-linear characteristics of an input variable, and extracting a main constituent from large to small according to contribution rates; then, processing a data set, and determining the correlation between an input vector and an output vector; and finally, performing modeling prediction, performing local weighing on a training sample and an input sample, and optimizing model parameters. The method adopts the defined health index, primary and secondary weighting matrixes can be built during a modeling process through the correlation between the input sample and an output sample and the distance between the input sample and a query sample; and the primary weighting can be performed before the sample is input to the model, the secondary weighting can be adopted to a model estimation error during the modeling process, and therefore, the influence brought by production data errors or omission on prediction algorithms can be reduced.

The invention discloses a multi-model space-time modeling method based on a finite Gaussian mixture model. The method is applied to a nonlinear distribution parameter system, a nonlinear space obtained by the nonlinear distribution parameter system is divided into a plurality of local operation subspaces based on a finite Gaussian mixture model, and an original complex nonlinear space-time dynamicequation is summarized into a plurality of simple nonlinear space-time dynamic equations, so that local modeling is carried out; when all the local space-time models are integrated, a principal component regression method is adopted to calculate the weight of each local space-time model, the existence of multiple colinearity is avoided, and a global space-time model of a large working area is reconstructed through multi-model modeling. The method provided by the invention has better performance for a large-scale, strong-nonlinearity and time-varying system.

The invention discloses a method for detecting honey sources. The method comprises the following steps of: (1) measuring a honey sample into a sample groove special for a rheometer, submerging a rotor of the rheometer into honey, setting the shearing rate range of the rotor and measuring the viscosity value of the honey by using the rotor; (2) adjusting the temperature of a rheometer water bath and measuring the viscosity values of the honey at different temperatures respectively by using the rotor; (3) establishing a main component analysis and clustering analysis mode recognition model by using the viscosity values of the honey sample at different temperatures to perform qualitative analysis on the honey sources; and (4) marking an encoding value for predicting respectively for honey ofdifferent honey sources and establishing a main component regression analysis predictive model by using the viscosity values of the honey sample at different temperatures to perform predictive analysis. During the implementation of the method, preprocessing is not required, so that the experiment steps are simplified, and the reliability and repeatability are enhanced greatly.

The invention discloses a roller kiln temperature soft measurement modeling method based on local twice-weighted kernel principal component regression. The method includes establishing a roller kiln temperature soft measurement model based on the local weighted kernel principal component regression through utilizing local sample data with high similarity, combining characteristics of the existinghigh dimensionality, nonlinearity and process time-varying of the roller kiln and introducing techniques of kernel tricks and instant learning separately; and performing twice-weighing on local modeling variables, and building the roller kiln temperature soft measurement model based on the local twice-weighted kernel principal component regression to achieve the precise prediction of the roller kiln temperature in consideration of the different influence degree of the input variables of local modeling sample data on the output variables. The obtained model can better track the state changes ofa process and provide good guidance effects for roller kiln temperature control, and therefore, product production quality and qualified rates can be enhanced.

The invention discloses a soft measurement method based on integrated orthogonal component optimized regression analysis, aiming at solving the problem that how to integrate and consider multiple types of orthogonal component regression algorithm and use the optimized idea to establish a soft measurement model. Specifically, the soft measurement method includes the steps: establishing three orthogonal component regression models by using a principal component regression (PCR) algorithm, an independent component regression (ICR) algorithm, and a partial least square regression (PLSR) algorithm;and then predicting the quality data again by using the predicted values of the regression models, wherein the difference is that the regression coefficient vector is obtained by using the particle swarm algorithm when the predicted values of the regression models are used to predict the quality data again. Compared with a traditional method, the soft measurement method based on integrated orthogonal component optimized regression analysis considers multiple orthogonal component regression models and predicts the output quality index through the optimized regression coefficient vector. Therefore, the prediction accuracy for the quality index, of the soft measurement method based on integrated orthogonal component optimized regression analysis is not less than any one regression model, andthe soft measurement performance is fully guaranteed.

The invention relates to a method for analyzing the influences of grain sizes on the magnetic performance of non-oriented silicon steel on the basis of principal component regression analysis. The method comprises the following steps: measuring the content of grains with different size ranges in the non-oriented silicon steel; standardizing all data; performing dimension reduction on the content data of the grains with different size ranges; calculating a characteristic value and determining a principal component and an expression thereof; performing regression analysis and performing significance testing on a regression equation; and converting the regression equation into a multivariate linear relation between the contents of grains with different size ranges and the magnetic performance of the non-oriented silicon steel by using inverse operation of a standard difference standardization method. According to the method, a multivariable program can be analyzed effectively, and the multivariate linear relation among variables is analyzed quantitatively by acquiring major information from complicated influence factors so as to quantitatively reflect the rule of the influences of the contents of grains with different size ranges on the magnetic performance of the non-oriented silicon steel, so that oriented guidance is provided for the practical production of efficient electrical steel products with low iron losses and high magnetic induction.

The invention discloses an electric vehicle charging safety early warning method. The method analyzes the consistency, performance degradation and safety characteristic coupling mechanism of a battery pack of an electric vehicle, and deduces the SOC consistency development rule of the battery pack; in order to further play the battery performance to the greatest extent and guarantee the safety of the battery pack, the safety state monitoring method of the battery pack is studied; in order to solve the multicollinearity problem in the power battery pack, the capacity of the lithium ion battery is estimated through principal component regression PCR, then regression is conducted on data through a multiple linear regression method, the result is compared with the result of the health state estimation model based on principal component regression, and finally, for short-term potential safety hazards, an outlier detection method is adopted to quickly and accurately detect an abnormal battery and give an early warning, and for long-term potential safety hazards, a power battery medium-and-long-term accelerated decline identification method is adopted to give an early warning. The method provides reference for safe charging of the electric vehicle.

The invention discloses a research method based on a GAMLSS model sediment transport contribution rate. The method comprises the steps of firstly collecting and arranging flow and sediment informationof a drainage basin outlet station; then computing a relation between each index and sediment transport and a fitting process line, computing 90% P-factor and R-factor, 50% quantile sequence varianceand a mean value, a correlation coefficient and an AIC; then using an attribution analysis method to analyze six indexes, thus acquiring an influence degree of each index on sediment transport variation; and at last, computing a contribution rate of each index to sediment transport. According to the research method based on the GAMLSS model sediment transport contribution rate provided by the invention, the problem that the prior art cannot accurately and comprehensively acquire the influence degree of the climate change and the human activities on the sediment transport contribution rate issolved, the principal component regression analysis is used, the influence of multicollinearity is eliminated, the influence generated by each variable mean is considered, and the influence of each variable interannual variance change on sediment transport is also considered.

The invention discloses a method for analyzing the influence of impurities on the magnetic performance of non-oriented silicon steel based on principal component regression analysis. The method comprises the steps that the number of the impurities in different dimensional sections in the non-oriented silicon steel is counted; all data are standardized; impurity statistical data are subjected to dimensionality reduction; a characteristic value is calculated, and the number of principal components and a principal component expression are determined; regression analysis is conducted, and a regression equation and an independent variable are subjected to a significance test; an inverse operation of the standard deviation standardization method is used for converting the regression equation into a multi-element linear relation between the number of the impurities in all the different dimensional sections, the total number of the impurities and the magnetic performance of the non-oriented silicon steel. By means of the method for analyzing the influence of the impurities on the magnetic performance of the non-oriented silicon steel based on principal component regression analysis, complicated variables can be involved in the operation with the same weight, influence caused by correlation between the variables is eliminated, the influence of the impurities on the magnetic performance of the non-oriented silicon steel is analyzed quantitatively, the dimensional sections of the impurities which affect the magnetic performance of the non-oriented silicon steel significantly can be distinguished visually, and directional guide is provided for actual production of electrical steel products with higher magnetic induction and low iron loss.

The invention relates to a method for quantitative analysis of multicomponent-adulterated pseudo-ginseng based on near infrared spectroscopy. The method comprises scanning an adulterated sample through a near infrared spectrometer to obtain a near infrared spectrum, respectively inspecting prediction accuracy and operation efficiency of principal component regression (PCR), support vector regression (SVR), partial least squares regression (PLSR), artificial neural network (ANN) and extreme learning machine (ELM) models, building an optimal modeling method and carrying out prediction on the content of the unknown sample through the optimal model. The method adopts a near infrared spectroscopy technology, is environmentally friendly and nondestructive and realizes fast detection. Through multiple multicomponent correction technologies and use of the optimal method, high accuracy is obtained. The method is suitable for quantitative adulteration of multicomponent-adulterated pseudo-ginseng.

The invention relates to a principal component regression analysis method for analyzing influence of texture components on magnetic induction of non-oriented silicon steel. The principal component regression analysis method comprises the steps of measuring contents of different texture components in the non-oriented silicon steel; performing standardized processing on all the data; performing dimension reduction processing on statistical data; calculating a characteristic value, and determining quantities of principal components and expressions of the principal components; performing regression analysis, and performing significance testing on a regression equation; if the regression equation has non-significant independent variables, performing significance testing on the independent variables; transforming the regression equation into a multivariate linear equation of magnetic induction about the contents of different texture components through inverse operation of a standard deviation standardization method. By means of the principal component regression analysis method, the multivariable problem can be analyzed effectively, multiple correlated variables can participate in operation with the same weight, quantitative research on the rule of influence of different texture components on magnetic induction of non-oriented silicon steel can be performed, texture components obviously significant in magnetic induction can be found out, and directional guidance is provided for actually producing non-oriented silicon steel products with excellent magnetic property.

The invention discloses a non-destructive testing method for plum hardness based on a visible/near-infrared spectroscopy. The method comprises the following steps: collecting different varieties of fresh plum samples, and using a hyperspectral image collecting system to collect hyperspectral images of plum samples; performing a black-and-white correction on hyperspectral images, and using median filtering and mathematical morphology processing method to construct mask images to extract an average spectral reflectance of the whole fruit region of plums; using an SPXY algorithm to divide the acquired spectral data into a calibration set and a testing set, preprocessing the spectral data by Standard Normal Variate (SNV) to obtain a spectral database of a calibration and test sample set; usinga digital fruit hardness tester to measure and calibrate and test the hardness values of the plum samples in the sample set; combining the Principal Component Regression method (PCR) with stoichiometry to establish a prediction model of plum hardness. The non-destructive testing method for the plum hardness based on the visible/near-infrared spectroscopy is capable of detecting plum hardness quickly and non-destructively based on the visible/near infrared spectroscopy.

Embodiments of the invention relate to sketching for M-estimators for performing regression. One embodiment includes providing one or more sets of input data. A matrix A and a vector b are generated using the input data. A processor device is used for processing the matrix A and the vector b based on a randomized sketching matrix S. A vector x that minimizes a normalized measure function is determined based on the matrix A and the vector b. A relationship between the input data is determined based on the vector x.