33 results about "F-test" patented technology

The invention discloses a method for forecasting the acute toxicity of organic compounds by building a quantitative structure-activity relationship model with a quantum chemistry method. The method fully geometrically optimizes compound structures by using a Gaussian procedure so as to obtain quantum chemistry parameters including molecular volume, relative molecular mass, highest occupied molecular orbital energy, lowest unoccupied molecular orbital energy, energy gaps of frontier molecular orbital, dipole moment, solvation energy, electron energy and the like; using the quantum chemistry parameters and a hydrophobicity parameter as structural descriptors; in combination with toxicity data, quantitative relationship equations between various structural descriptors and toxicity are established according to a written procedure based on partial least square stepwise linear regression to obtain the multiple correlation coefficient, F-test value and sum of squared residuals, and then the model is verified so as to guarantee the external predictive ability. Therefore, the method can quickly and effectively forecast the toxicity of organic compounds to be studied, and provide necessary basic data for risk assessment and supervision of chemicals.

The invention discloses a method for modeling the location-dependent geometric error item of a numerical control machine tool, which belongs to the technical field of numerical control machine tool machining. The method comprises the following steps: determining the n-order polynomial form of the location-dependent geometric error item of a numerical control machine tool according to the property of the location-dependent geometric error item; according to the discrete data of the location-dependent geometric error item of the numerical control machine tool, getting an n-order polynomial coefficient calculation matrix M based on the principle of least square fitting, and calculating the coefficients of the n-order polynomial; calculating a series of polynomials from first-order to mth-order polynomials according to the discrete data of the location-dependent geometric error item of the numerical control machine tool, wherein the highest order m is set to half of the discrete data of the geometric error item; calculating the F values of first-order to m-order polynomials through a regression analysis F-test method; and finding out the maximum F value through comparison, and selecting the polynomial corresponding to the maximum F value as an optimized polynomial model of the location-dependent geometric error item of the numerical control machine tool. The method is mainly used to realize automated and programmed machine tool comprehensive error modeling.

The invention provides a forecasting method for the surface roughness of strip steel in the rolling process and belongs to the technical field of plate strip rolling. According to the method, on the basis of batch collection of industrial production site data, data of the initial surface roughness of a working roller, the surface roughness of the strip steel and production process parameters of the strip steel are obtained; by adopting a regression analysis method, the regression coefficient is calculated by utilizing a least square method; and independent variables with insignificant influence on dependent variables are eliminated one by one through a F test, and accordingly a forecasting model of the surface roughness of the strip steel is obtained. According to the forecasting method, considered process factors are more comprehensive, the principle is simple, the calculation process is convenient, and actual site process parameters can be fully utilized.

The invention discloses a small sample data model verification method based on statistical analysis, relating to a small sample data model verification method. The invention aims to solve the problemsthat the scope of a conventional Bootstrap method for reproducing samples is limited to an original sample range, especially in the case of a small sample size, the distribution of the reproduced samples may deviate from the real distribution, the estimation results may be inaccurate, and certain risks exist. The method includes the following processes: step I, performing a normality test on a reference sample and a simulation sample, and if obeying the normal distribution, performing step II; and step II, when n is greater than or equal to 30, adopting a U test method; when n is greater than10 and less than 30, adopting a t or F test method; when n is greater than 3 and less than or equal to 10, adopting a formula 1 and a formula 2 (as shown in the original specification) to separatelyperform a single normal population parameter test on the simulation sample in the step I; determining whether the obtained mean value and variance of the reference sample and the simulation sample areconsistent; and when n is less than 3, not performing model verification. The scheme of the invention is applied to the field of simulation model verification.

The invention discloses a method for predicating the acute toxicity of organophosphorus pesticide on aquatic organisms through the quantitative structure activity relationship. On the basis that the structure of chemical compound is known, geometrical full optimization is performed on a Gaussian program to obtain the MV, the MW, the EHOMO, the ELUMO, the difference value delta E (hartree), the DM and the like to serve as structural descriptors, regression parameters of all parameters, the multiple correlation coefficient and an F-test value are obtained in combination with data of toxicity of the organophosphorus pesticide on aquatic organisms through the programmed partial least squares stepwise linear regression program, and quantitative relationships between the various structural descriptors and the toxicity. Thus, the toxicity of the organophosphorus pesticide on the aquatic organisms can be fast, conveniently and effectively predicted, and essential basic data are provided for risk assessment and supervision on the organophosphorus pesticide.

Embodiments of a system, method and apparatus incorporate two-stage identification of targets in hyperspectral image analysis. In various embodiments, unmixing is employed that integrates F-test and model averaging approaches. Further, a multi-tier target library process provides an improvement in the spectra that can be used to detect target materials and spectra that can be used for unmixing in identification. Additionally, the hierarchical identification of the present disclosure combines probabilities from model averaging to generate target identifications simultaneously at multiple levels of specificity.

An LS (least square)-based multi- model adaptive state estimation method includes the following steps: S1, same moving object observation sequences of sensors are acquired to form moving object X, Y and Z direction observation sequences in the space rectangular coordinate system at different times; S2, the degrees and the number of time-variable polynomial equations are determined, and LS (least square)-based polynomial equation coefficient estimation of the moving object X, Y and Z direction observation sequences is performed respectively; S3, according to the polynomial equations in the step S2, regression sum of squares and residual sum of squares are respectively calculated for F-test to determine a best fit polynomial; S4, a corrected best fit polynomial is determined; S5, a first time partial derivative and a second time partial derivative are sought respectively to obtain moving object corresponding velocity and acceleration polynomials; and S6, object position state estimator coordinate conversion is performed to obtain corresponding position coordinates, and then by combining with velocity estimated values, object state estimation can be completed.

The invention relates to a preparation method and application for an electrochemiluminescence sensor based on protein activity protection and belongs to the technical field of electrochemiluminescencedetection. An efficient electrochemiluminescence behavior capable of being achieved by europium-doped gadolinium phosphate through excitation under low potential is developed and is verified for thefirst time. The fixing problem of a luminescent material on an electrode is solved. The effective preservation problem of antigen-antibody activity is also solved. Concentrations are detected according to the fact that intensities of electrochemiluminescence signals responding to procalcitonins of different concentrations are different. Through F-test and T-test, accuracy and precision of the method are shown. Test results are smaller than a theoretical value. The method is accurate and reliable.

The invention provides a fault diagnosis method and device for an electro-hydraulic servo valve, a computer readable storage medium and electronic equipment. The method comprises the following steps:acquiring data of a plurality of electro-hydraulic servo valves; dividing an obtained characteristic curve into an upper half part characteristic curve and a lower half part characteristic curve by taking a flow value as an boundary; calculating a flow difference value of the upper half part characteristic curve and the lower half part characteristic curve, which correspond to the same current value; adopting a Kolmogorov-Smirnov method to verify whether each flow difference value meets Gaussian distribution or not; if each flow difference value meets Gaussian distribution, calculating a meanvalue and standard deviation of the flow difference value of the upper half part characteristic curve and the lower half part characteristic curve of each electro-hydraulic servo valve; separately performing statistiacal verification on each electro-hydraulic servo valve by F-test based on statistics, and judging whether each electro-hydraulic servo valve has faults or not. The method and the device realize efficient and accurate diagnosis for faults of the electro-hydraulic servo valves, and provide large data basis for electro-hydraulic servo valve fault identification.

The invention relates to a hypothesis testing method for sensitive parameters of a fire work system. The method comprises the following steps: (1) selecting the sensitive parameters of the fire work system; (2) counting historical sample data of sensitive parameters of the fire work system; (3) carrying out normality test and transformation on the historical sample data of the sensitive parametersof the initiating explosive system, so that the historical sample data of the sensitive parameters of the initiating explosive system and the to-be-tested sample data are in normal distribution; (4)when the number of historical sample data of the sensitive parameters of the fire work system is greater than or equal to a preset threshold, verifying significant changes by adopting a single normaltotal mean value t test method; and (5) when the historical sample data volume of the sensitive parameters of the fire work system is smaller than a preset threshold, verifying the significance changeby adopting a two-normal overall mean value t test method, and if the two groups of sensitive parameter samples do not have the significance change, verifying the significance change by adopting a two-normal overall variance F test method. According to the invention, more accurate inspection is realized.

A performance degradation feature assessment method based on time series change mutation point detection comprises the steps of: providing four performance degradation feature assessment criteria consisting of an initial degradation point, the initial sensitivity, the failure mutation and the trend consistency; for the four performance degradation feature assessment criteria, respectively constructing quantitative indexes; constructing a linear regression equation of state variables and time series; solving coefficients of the linear regression equation, and extracting time series turning points by employing the F test; determining starting and ending points of the degradation state change, and calculating performance degradation feature assessment indexes; and finally, comparing indexes of different performance degradation features to assess the performance degradation features. The performance degradation feature assessment method provides the performance degradation feature assessment criteria and the corresponding quantitative assessment indexes having guiding significance on the extraction of the performance degradation features, greatly reduces the computational complexity and is high in efficiency.

The invention belongs to the technical field of finished automobile information and entertainment systems, and relates to an objective quantification method for subjective evaluation of a warning property of an automobile warning tone. The method includes the steps that a multiple linear regression model is established, a subjective evaluation performance value is taken as a dependent variable, an acoustic parameter value is taken as an independent variable, the model is alpha = a * L + b * Sh + c * F + d, and in the formula, alpha is the subjective evaluation performance value; l is the loudness, Sh is the sharpness, and F is the frequency of the sounding interval, and a, b, c and d are undetermined coefficients; regression analysis is performed by adopting a step-by-step method; independent variable selection is performed according to a rule that a variable with a P value of less than or equal to 0.050 enters and a variable with a P value of more than or equal to 0.100 is shifted out according to an F test; acoustic parameter values are input into a formula for subjective evaluation performance value calculation, and when the prediction error of the subjective evaluation performance value is within 10%, it is considered that the multiple linear regression model has the effect of predicting a subjective evaluation result through automobile prompt tone objective test data. The influence of human factors on the evaluation result is reduced, the discreteness of the evaluation result is reduced, the evaluation efficiency is improved, and the authority of evaluation is increased.

The invention discloses a method for determining a causal relationship between a water resource and an ecological environment based on a Granger test. The method comprises the following steps: S1, obtaining a water footprint time sequence and an ecological footprint time sequence; s2, carrying out stationarity test on the two time sequences; s3, carrying out co-integration on the two time sequences, and judging whether the two time sequences have an equilibrium relationship or not; s4, establishing a coupling model of the water footprint and the ecological footprint, and calculating a coupling degree value of the two time sequences by using the coupling model; s5, performing Granger causal relationship verification on the water footprint time sequence and the ecological footprint time sequence; s6, constructing a regression model, and calculating an F statistic by using the regression model; and S7, determining the causal relationship between the water resource and the ecological environment according to the F statistic in combination with an F test critical table. According to the method, the causal relationship among the data is determined by utilizing Granger causal relationship verification, the directivity is clear, the constructed relationship accords with reality, and the accuracy is improved compared with the existing scheme.

The invention relates to an elasticity modulus testing method based on a virtual pure bent material, and belongs to the technical field f test or analysis of materials by applying stable tension or pressure to test physical or chemical shapes such as strength character of the solid material. A long and thin component made of the material is used for carrying out a free vibration mode test, a corresponding virtual force vector is constructed according to support constraint conditions of components, a flexibility matrix of the component and the virtual force vector are multiplied to obtain a displacement parameter under a virtual pure bent state, then the curvature of the virtual pure bent section is calculated, and finally, the equation is listed according to balance conditions of the crosssection of the virtual pure bent section to obtain elasticity modulus. The elasticity modulus testing method based on the virtual pure bent material is applied to existing components which have various support constraint conditions and cross section types and have been used for a long time, and has the advantages of simplicity in operation, short time required for test, zero damage, zero limitation to an operation environment and the like.