Corrosion State in Metal Pipes and Fittings Exposed to Air Based on Fuzzy Sets

Abstract

Metallic materials used in drinking water distribution systems show corrosion damage, which makes it necessary to develop simple and accessible methods that allow preventive and / or corrective measures to be carried out quickly. The work that is described in this paper intended to build a fuzzy logic model based on data obtained from visual inspections of the cladding, and the visible deterioration appearances degree (aging) of the elements. The methodological design used was quantitative research, also known as a mathematical approach, distinguished by the interpretation of data, statistics, numbers, and tables, among others. The data and records of various pressure reducing stations (ERPs) in zones 1 and 4 of aqueduct and sewer in Bogotá were analyzed. Fuzzy Logic ToolboxTM MATLAB software complementing was used, based on the selection of two parameters (protection, and coating and aging). Each one was assigned quality conditions that operated as the program input data. By applying the model, a three-dimensional surface was obtained with the values of the corrosion states, according to the quality of the coating and the aging dimension. Four zones were identified, which represent the corrosion states of the evaluated elements: optimal, imperceptible, light, and severe, with no transition interval. The results show that this methodology provides a reliable evaluation of the state of corrosion of the evaluated metal structures. It is concluded that the implementation of the fuzzy inference system provides a simple application tool, which can be adjusted with information collected in routine preventive or corrective maintenance work, and in activities that relate other variables to corrosion.