The Apply Research Of Multivariate Statistical Analysis In The Processing Of Materials Test Data

The article based on the multivariate statistical analysis, making the orthogonal test design of mixing the quantity proportioning of concrete take the high pulverized coal ash as the example, making the uniform experimental design of duplicate mixes the configuration high performance concretes take the mineral substance fine dust as the example, from the experimental design stage, using the concretes cementitious material, fine aggregate, coarse aggregate, water-cement ratio factors and so on to the concretes mechanical properties' influence, using the multivariate linear regression analysis method to optimize the calculation of the regression program,select optimization plan, fit equation.Application range and variance analysis and F values test were used parametric regression analysis, observing the instrumental accuracy and prediction error's size, simultaneously, applying the multi-dimensional quality control chart to carry on the multivariate outliers, analyzing the fitting equation reliability. Using combined equation proposed, analysis the mechanical properties and so on preliminary relations of concretes compressive strength parameter. Making use of the variable relation that equation reveals, design and manufacture of commodity concrete being able to guide from physics and chemistry parameter to improve the mechanical properties of concrete and other building materials as well as economy is constitutive purpose.This article applies the multi-dimensional statistical analysis to divide into three steps to carry on the application elaboration. The first step is the factor choice, including quality objectives determination and experimental factor choice. The second step is the experimental design and the model building. According to the actual situation arrangement experiment, and establishes the corresponding model.Discussing the orthogonal polynomial regression in complex mixed high performance concrete application, and based on normal concrete mixing of fly ash, silica fume and other mineral powder,presenting model of high-performance The mechanical concrete, and makes further amendments to the model The third step is quality control and validation, multivariate statistical methods such as principal component analysis and factor analysis to inspect the validity of the conclusions of test model.