Strain of heated concrete during two thermal cycles. Part 3: isolation of strain components and strain model development

摘要

This paper follows on from the two previous papers in the series dedicated to the strain behaviour of three nuclear reactor type concretes during a 14-day two-thermal cycle. While the trends were described in detail in the previous papers, the present paper is dedicated to the separation and quantification of strain components during the various stages of the heat cycles in order to develop a predictive strain model for use infinite element modelling. The total strain is shown to be the superposition of individual strain components, each related to the specific strain inducing mechanism within the concrete. Having established the principles of strain components, the strains measured during the first thermal cycle under load were isolated, and quantified, according to the strain type and heating stage: (a) virgin heating under load: load-induced thermal strain (LITS) and shrinkage components; (b) constant temperature: creep, shrinkage and crack-induced strain; and (c) cooling: contractive thermal strain, plus chemically and crack induced expansive strains. The methodology used required a number of specimens (S~0, S~σ). Isolating the shrinkage component at temperatures below about 400℃ required only one specimen. LITS required two specimens: one loaded and one unloaded. With this methodology, the strain components were isolated and quantified. The test for prediction by the strain model was made against the residual strain from one specimen whereas the individual strain components were derived from several tests. The master LITS component was evaluated separately in a generic sense. The residual strains consist of the sum of the irrecoverable contractive and expansive strains that occur throughout the thermal cycle, the initial elasto-plastic strain having been removed from the outset. The prediction was successful for test temperatures (i.e. maximum temperatures) of up to 400℃ without considering a crack-induced component. The average value of the ten predictions taken for test temperatures between 110 and 400℃ for the three concretes is 100.3%, thus justifying this method of prediction. Above 400℃, the residual strains predicted for the three concretes were 83, 93 and 98%. These undershot the actual measured value, thus confirming that applied load reduces the crack-induced component thus requiring another term (ε_(tr, crack-sup)~(σ, T, d)). Also the degree of 'undershooting' is related to the amount of cracking in the concrete and its thermal stability. Here again, the results confirm that the thermal stability increases in order for the limestone, basalt Ⅰ and basalt Ⅱ concretes.