Targeting Colour and Modulus of Resistance for a repointing mortar through a genetic algorithm

Abstract eng:
Smart construction systems have been used for centuries in many places around the world, masonry is widely encountered in numerous heritage structures to be restored. For durability purposes, preservation operations that will be carried out on such structures require achieving a sufficient compatibility between parts that will be conserved and materials that will be brought through the restoration process. Classically, incompatibility problems appear in relationship with the formulation of the mortars that will be used for repointing or repairing works. For setting up the compatibility problem, engineers usually focus on three axes: the mechanical one (Modulus of Resistance and/or Modulus of Elasticity), the physical one (gas and/or liquid transfer capabilities) and the perceptive one (colour and/or texture). In the daily practice, some practitioners typically insist on aesthetical aspects as they intend not to alter the perception of treated buildings. Replicating a satisfying texture is achieved by acting on sand granularities and finishing tools while the colour question usually requires manual trial and error procedures to be followed. Other practitioners privilege mechanical aspects for avoiding progressive damages induced by stress concentrations and target suitable values of MOR and MOE by relying on empirical recipes. Practically gathering both the approaches inside a same project is rare because it reveals difficult. The present paper describes an automatic computer-aided approach relying on evolutionary strategies (elitist genetic algorithm) in order to establish a mortar formulation encountering both aspects. The global framework in which the reflexion may take place, the optimization problem as well as the sharp manner to take each aspect into account are detailed. In its preliminary shape, the proposed tool focuses on the integration of strength and colour aspects but the method is designed to be extensible to complementary characteristics in a near future.

• Export as: BibTeX | MARC | MARCXML | DC | EndNote | NLM | RefWorks
• Add to your basket