br Results and discussions In general experimental

Results and discussions In general, experimental analysis is time consuming and expensive. Thus, In the present study, Taguchi design of experiments (DOE) was adopted to reduce the number of trials. To analyze the data, a High Level Analysis (HLA) is performed by averaging measured data for each level of a single parameter, then plotting the averaged data against all the levels of that parameter. HLA has high statistical significance. In the present data analysis, effect of each parameter is calculated by averaging nine trials containing the corresponding parameter in the data points. The experimental design and results are shown in Table 1. The effect of the individual parameters is analyzed below.
Development of an empirical model for RHA replacement in cement concrete to enhance its strength and durability In the present experimental study, a large number of experiments were performed to determine the strength of the cement concrete. During these investigation more number of factors were considered and employed over these experiments. Hence, given the handiness of large number of experimental data, it turns possible to develop a model. This model can predict the relationship between RHA added cement concrete mechanical strength (MS) namely compressive strength and tensile strength, durability (D) of the same while subjecting to corrosive environment (5wt% H2SO4 and 5wt% Nacl solution) and the factors that are influencing the cement concrete. The factors that considered for developing this model are RHA loading, RHA initial (Dpi) and final size (Dpf) (size reduction ratio), bulk density of RHA (ρ), specific surface area of the RHA (Asp), pozzolanicity or chappel activity (CA), finness of modulus (FM), loss on ignition (LOI), normal curing time (TNC) (7, 14 and 28days), TUG 891 attack (TAc) and alkaline attack (TAlc) curing time, silica (SiO2) content of RHA and most importantly thermal diffusivity (α). The same is given in the form of Eq. (4.1). Hence, the dimensional analysis was performed to identify the relevant non-dimensional groups. The general form of the developed cement concrete mechanical strength and durability (Acid and alkaline attack) equation are given in Eqs. (4.2), (4.3), (4.4).where the K1, K2 and K3 are empirical constants and a, b, c, d, e and f are the model parameters. The left hand side of the above equation refers the ratio between compressive and tensile strength of the cement concrete. Then, the right hand side of the above equation containing six number of non-dimensional groups having different physical significance. The first group indicates the relation between RHA and its silica content, the second and third refers the RHA particle size reduction ratio and their surface area and closeness of the RHA particle packing in the form of bulk density. The fourth non-dimensional group belongs to the curing period and the thermal diffusivity. Finally, the fifth and sixth groups narrates the Chappel activity (CA) and the relation between fine modulus and loss on ignition of RHA. In durability prediction Eqs. (4.3), (4.4), the fourth non-dimensional group is different from the mechanical strength prediction Eq. (4.2). The reason was that the test coupons are water cured for 28days prior to 90days of acid and alkali soaking hence it is included in the model. Large number of experimental data were used to develop these correlations between mechanical strength and durability of the RHA added cement concrete. A least square method was employed to arrive the empirical constants and model parameters with higher accuracy. And the final models are given in the following Eqs. (4.5), (4.6), (4.7). From the above three equations, it has been observed that the size reduction ratio, pozzolanicity, specific surface area, curing time and the thermal diffusivity are the stronger functions of determining the strength of RHA added cement concrete. The remaining factors like RHA loading, silica content and the ratio of FM and LOI becomes secondary since the individual effect of silica content in RHA plays a vital role in strength development. In addition, the silica content of different types of RHA used in this study were insignificant among them. Fig. 10 shows the comparison between the measured values of compressive strength, compressive strength after acid and alkali attack, and the values predicted by the model represented in Eqs. (4.5), (4.6), (4.7). The model appears to be able to predict the mechanical strength and corrosive resistance of the RHA added cement concrete quite accurately with >90 confidence. Model based trial concrete mix can be performed with highly parameterized developed accurate empirical models, particularly when predicting highly nonlinear responses such as concrete strength under various corrosive environment and various % of RHA replacement in concrete. In an alkaline and acid environment certain properties like magnitude of corrosive nature of the environment, % RHA replacement, particle sizes and curing period needs to be carefully ascertained in concrete making process in order to develop the strength of concrete that are implemented in this model. Since the model comprising of numerous important influencing factors furnishes the convenient way of screening the high strength concrete mix trials.