Aspects regarding the prediction of earth electrode corrosion in the soil of Timisoara

. Corrosion prediction of galvanized steel ground electrodes in the soil of Timisoara represents an area of interest for ensuring the good functioning of the earth electrodes , the safety of the electrical installation, and of the lightning rodes. Taking in the account that the soil is exposed to different kind of pollutants which have an effect on the ground electrodes we consider that a research is needed in this area.


Introduction
The city of Timisoara is located in the Banat Plain in western Romania, and has the following geographical coordinates: latitude 450 47' 58'' N and the longitude 210 17' 38'' E [1]. Historically, the city is mentioned documentary in the year1212 or 1266, but the hearth of the city settlement attest to the presence and human activity since the neolithic period, due to the high fertility of the soil and a military strategic position, being surrounded by marsh and several water channels. The soil [2] is of type: chernozem, gleysol, leachate, reddish brown. And local soils: swamps and solonetz with medium-fine texture, low -medium porosity, and low permeability.
The definition of earthing installation: the earthing installation is the assembly of electrical conductors through which the link with the soil is secured [3]. In a building, the connection of every metallic part of the building, and the exposed conductive parts of the equipments to the building earthing socket prevents the accidental occurrence of dangerous voltages between two metallic parts simultaneously. [4] The earthing installation ensures the protection of electrical equipments and the building itself against lightning strikes, the limitation of electromagnetic perturbations, and the safety of humans and animals by limiting the pitch voltages [5]. In Cap.5 from Indicativ I7/2011 [6] subcap.5.5, Sisteme de legare la pământ pct 5.5.7. The earth socket is made of one or more electrodes (earth electrodes The ones selected in this study are: total porosity[%] (capillary porosity and non-capillary porosity (of aeration)) and the hygroscopic coefficient[%], as elements that lead to the corrosion of earthing electrodes and their connections. The element which determines the electrical resistance of a earthing socket (besides the length of the electrodes) is the resistivity ρ of the soil. The soil is not homogeneous, and the deeper layers have a more stable and lower resistivity. The resistivity ρ characteristics of the soil are presented in Table 2, Table 3, Table 4 and Table 5.  Resulting that sand and argil (pottery clay) predominate the soil composition of the sample no.1(a)and(b) 3 Materials-specimens of steel used for obtaining the earth electrodes An important role of the earth socket it represents the resistance RL, respective the metallic part of the earth socket: vertically positioned earth electrodes, and the metal strip horizontally positioned both in the ground and above ground. As materials used for making the earth electrodes, we mention: -Corner profiled steel, T, I, and the metal strip unprotected; -Corner profiled steel, T, I, and the metal strip protected through galvanizing; Corner profiled steel, T, I, and the metal strip protected with bentonite; -Unprotected steel pipes; Galvanized steel pipes; -Round steel unprotected and galvanized; -Round steel protected with bentonite; -Round steel unprotected and embedded in concrete -Copper (pipes, plates, band); -Stainless steel. In order to perform the corrosion measurements of the earth electrodes, the materials presented in Table 6 were used.  The Tafel equation was first experimentally deducted, and later proved to have a theoretical justification. The equation is named after the swiss chemist Julius Tafel) [15]; Gravimetric method with high and low humidity (Gravimetric analysis, also called gravimetry [1] , brings together all analytical methods used in analytical chemistry for the quantitative determination of an analyte (the compounds of interest) based on its mass. [2] The principle on which gravimetric analysis is based is the mass determination of a compound derived from the sample analyte, and on the basis of the chemical reactions to which it has been reached, the mass of the respective analyte can be determined. For this reason it is necessary to know in detail the chemical processes that take place, and these must be total (quantitative)) [16].

23÷49
20÷36 15÷17 The soil sample that was used for the experiment is theArgil (pottery clay)(According to the specialty literature with the resistivity ρ=40Ωm, considered a poorly aggressive environment, and ρ=20Ωm, considered moderately corrosive environment). The results of the analyses carried out by the two methods are presented in Table 9 and Table 10.  In Figure 1 we can see the soil classes existing on the Romanian territory. Figures 2 to 4 show images taken during the measurements.

Corrosion of steel and copper specimens
Procedure: The specimens were placed in the DCTC (Dry Corrosion Test Cabinet) and exposed to multiple spray processes, followed by a rest period. A cycle is constituted of a 5 minute spray step followed by a 55 minute rest period. In total, the DCTC test lasted 60 days, with 1440 cycles. The saline solution was prepared by dissolving 1 kg of sodium chloride in 20 L of distilled water. The temperature inside the test cabinet was 25 ± 3℃, both during the spraying process and the rest period. The temperature inside the test cabinet was 25 ± 3℃, both during the spraying process and the rest period. The temperature inside the test cabinet was 25 ± 3℃, both during the spraying process and the rest period. The humidifier temperature was 25 ± 3℃. The spraying pressure was 1.4 bars, and the sprayed quantity was of 80 mL/h * m3. The two specimens were turned every 72 hours. The pH of the saline solution after preparation was 7.37, and 8.54 at the exit of DCTC.

Corrosion of the Cu specimens
Two plates and two bars of Cu -of which a plate and a bar were to be put inside DCTC, and the other plate and the bar served as witnesses -they were weighed before and after the cleaning process. The cleaning process of the Cu plates and bars consisted in: Washing with detergent; Rinsed with water and wiped; Washing with a solution of HNO3 5%; Rinsed with water and washed with acetone; Dried in the oven at 40℃ for 60 minutes.

Corrosion of the galvanized steel specimens
The corrosion study of the galvanized steel specimens was also achieved through the effect of the salt mist, starting from 20.06.2018 until 15.07.2018.

Mathematical prediction
In statistics, regression validation is the process of deciding whether the numerical results quantifying hypothesized relationships between variables, obtained from regression analysis, are acceptable as descriptions of the data. The validation process can involve analyzing the goodness of fit of the regression, analyzing whether the regression residuals are random, and checking whether the model's predictive performance deteriorates substantially when applied to data that were not used in model estimation. One measure of goodness of fit is the R 2 (coefficient of determination), which in ordinary least squares with an intercept ranges between 0 and 1. While a low R 2 implies that the model does not fit the data well. In statistics, the coefficient of determination, denoted R2, is the proportion of the variance in the dependent variable that is predictable from the independent variable(s). In regression, the R2 coefficient of determination is a statistical measure of how well the regression predictions approximate the real data points. An R2 of 1 indicates that the regression predictions perfectly fit the data. In regression, the Null Hypothesis is that the coefficients associated with the (independent) variables is equal to zero. The alternate hypothesis is that the coefficients are not equal to zero (i.e. there exists a relationship between the independent variable in question and the dependent variable). When p Value is less than significance level (< 0.05), we can safely reject the null hypothesis that the co-efficient β of the predictor is zero.

Conclusion
In conclusion, a regression model is appropriate if the Pvalue of the test t is small (below an error threshold of 0.05) and the value of the coefficient of quantification as high as possible (closer to 1).