||High performance concrete (HPC) is used in more durable and serviceable
structures, such as high rise buildings and highway bridges. In a lack of a
generalized systematic HPC mix proportion procedure, numerous trial mixes
are required before a suitable mix proportion is decided. Therefore, there is a
need to have a generalized HPC mix proportioning procedure that can be
used to get satisfactory mixture with desired concrete performances, in order
to minimize the number of trial mixes.
A fuzzy adaptive resonance theory (ART)-based HPC mix proportion
estimation model can be used as an aid to HPC mix proportioning to get a
satisfactory mixture which can achieve desired concrete performances and
minimize the number of trial mixes. The fuzzy ART-based high performance
concrete mix proportion estimation model is a model that is able to estimate
HPC mix proportion that satisfies a set of desired concrete performances,
based on real experimental data. The model receives a set of desired
concrete performances, searches for a set of mix proportions which satisfies
the desired concrete performances, classifies the mix proportions of selected
data into clusters, measures the similarity between performances of deduced
clusters with desired performances, and deduces a mix proportion. As the
model uses clustering method, it is able to identify potential noisy data which
might affect the output of the model.
Two additional features of the model is its ability to:
(i) handle multiple desired concrete performances, and
(ii) to propose mix proportions with different concrete constituents.
This model is useful to aid HPC mix proportioning process, in order to get a
mix proportion that satisfies a set of desired concrete performances.
Experimental works have been conducted to validate the output of the model.