Application of the Method of Multivariate Analysis to Assess the Quality of Coatings

Information is given on the application of multivariate statistical techniques for analyzing the causes of the declining quality of the coatings and the occurrence of defects. It is shown that the porosity of the surface is a decisive factor in determining the quality of the coating.


Introduction
Previous studies show that the resistance of coatings, among other factors determined by the quality of appearance of coatings [1,2,3].The quality of the appearance of the coatings are strongly influenced by the technology of applying paint, its rheological properties, the quality of the surface.It is known that any process is subject to variation, the nature of which is determined by the influence of a set of random and non-random factors.These include variability in feedstock from party to party, wear process equipment, inadequate technological methods, different qualifications and other performers.
Earlier studies show that the process of creation of coatings on porous cement substrate is often unstable and irreproducible [4,5,6].In this connection is urgent assessment of the most significant factors affecting the quality of the coatings.

Methodology of the research
In this paper we present the results of the evaluation of the possibility of multivariate statistical analysis to assess the causes of the most significant factors reducing the quality of the coatings and the occurrence of defects.The method of canonical correlation allows to simultaneously analyze the relationship of several output parameters and a large number of determinants.Algorithm of calculation method of canonical correlation is constructed in such a way that the original variables are replaced by their linear combinations.The coefficients in the canonical variables characterize the effect of influence factors relevant traits At the same time a high degree of connection between the linear combinations of the factors and linear combinations of the output parameters.
Briefly present the essence of the method.Primarily through the array of measured values is calculated covariance matrix of factors reflecting the statistical picture of the state of the process:   To analyze the influence of the method of application of the paint composition, its rheological properties and quality of the substrate on the surface quality of the coatings we had to do the following experiment.
Colorful compositions with different rheological characteristics of the mortar applied onto the substrate porosity of 24%, 28%, 32% in two layers with intermediate drying for 20 minutes.Before applying the colorful composition of the surface of the substrate priming.In addition, part of the mortar samples leveled spackling compounds.The rheological properties of paints were evaluated in terms of their conditional dynamic viscosity and surface tension.As colorful compositions used alkyd enamel brand ПФ-115, oil paint brand MA-15, an acrylic latex (facade) paint.Colorful compositions were applied pneumatically, brush.The surface quality of the coatings was evaluated in terms of roughness and adhesion of coatings.The surface roughness of the coating was determined by the device profiler mark TR-100, the adhesion strength -the pull washers.

The research results
Analysis of the data presented in Table 1 indicates that the value of the surface roughness of the coating depends on the application of the paint composition, rheology of the cement and the porosity of the substrate.Thus, for oil paint MA-15 (green color) the minimum value of surface roughness Ra = equal 3,12mkm achieved on a substrate with a porosity P = 24% when the ink viscosity 0,00261*10 3 Pa*second when applying it with a brush.To paint ПФ-115 minimum value of roughness Ra = equal 1,3mkm achieved on a substrate with a porosity P = 28% when the ink viscosity 0,00065*10 3 Pa*second when applying it with a brush.For latex paint minimum roughness value equal to Ra = 3,5mkm achieved on a substrate having a porosity P = 32% when the ink viscosity 0,013*10 3 Pa*second when applying it with a brush, and a maximum roughness value equal to Ra = 6,5mkm achieved on the substrate with a porosity P = 24% when the ink viscosity 0,0347*10 3 Pa*second when applying it with a brush.The minimum value of the roughness characteristic of the surface coating formed on the substrate, the filler irrespective of the method of application and the rheological properties of colorful compositions.Since the dispersion factor variables are significantly different from one another and are dissimilar units, it is reasonable to use the correlation matrix for the preparation of which apply the Table 2.  .Thus, the contribution factor of the first variable in the overall instability of quality indicators is greater by more than 4 times greater than the contribution of the second factor.

SSR
Then the matrix S is represented as a block matrix, combining the characteristics of individual blocks of variations of factor variables, result indicators and their pair wise combinations.In fact, we divide the matrix S into four parts: where 11 S -the covariance matrix of the factors represents the transpose of the matrix 12 S .Canonical correlation coefficients can be calculated and based on the sample correlation matrix, especially if you have to work with these disparate units of measurement.The problem of determining the maximum correlation between the canonical variables

Table 2 -
Factors of variation of the experiment