Estudo do comportamento mecanístico-empírico de materiais alternativos e cimentícios em camadas de pavimentos

Abstract

The increasingly austere environmental policy, associated with the decrease in supply and the increase in the costs of natural resources used in the construction of road infrastructure, leads researchers to study the technical, environmental and financial and/or economic viability of waste materials and mixtures stabilized as raw material for sustainable application in engineering works. The motivating challenge of this research is to contribute to the study of the behavior of various materials (soil, thermoelectric ash and stabilized ash-lime and soil-cement mixtures) for paving purposes, through the design and execution of an Experimental Section with ash; the creation of a Manual for the Application of Ash in Paving; the proposition of models for predicting permanent deformation (DP) in soils; and the study of the mechanical behavior of soil-cement mixtures in light of mechanistic properties, such as DP and fatigue. To this end, the thesis was organized into 8 chapters, Introduction, Theoretical Framework, Article 1, Article 2, Article 3, Article 4, Article 5 and Conclusions. The research included all the physical and mechanical tests that resulted in the proposal of the structural pavement project for the section, as well as the stages relevant to the execution, observing particularities of the material worked on, which will be documented with the aim of assisting the applications of the Ashes on the pavement. The manual of good practices for the execution of pavement layers with coal ash developed practical guidelines for the use of ash in paving, proposing usage methodologies and technical service specifications for the use of this material in the road area. The soil PD prediction model in the Metropolitan Region of Fortaleza was created using the Artificial Neural Networks technique, with the California Support Index (ISC), confining tension, deviation tension, Resilience Modulus ( MR), AASHTO classification and number of cycles (N). The final model has an architecture formed by 6 neurons in the input layer, 20 neurons in the first intermediate layer, 10 in the second intermediate layer and 1 neuron in the output layer (6:20:10:1). In relation to cemented mixtures, the increase in cement content and curing age in soil-cement mixtures promoted an improvement in the resistance and stiffness parameters in the tested samples. With regard to the Permanent Deformation parameter, a tendency for plastic deformations to stabilize after 6000 load application cycles in the triaxial test was observed, confirming this trend through analyzes in the MeDiNa software. In terms of fatigue, fatigue life models of cemented mixtures were generated, carrying out mechanisticempirical analyzes of the behavior of the mixtures investigated as the base and sub-base of road pavements. Therefore, the studies developed here can guide new projects that choose to use waste materials from thermoelectric plants and soil-cement mixtures in pavement layers, enabling better prediction of pavement performance based on analysis using more mechanistic routines.