| Abstrakt: |
Concrete compressive strength is the most important performance requirement in structural engineering when developing both traditional concrete and high-performance concrete (HPC) constructions (CCS). When using this parameter in time and cost–benefit analyses, it is much more important to forecast it correctly. Because of the complexities of the development processes, the compressive strength of concrete is highly sensitive to a variety of factors and notoriously difficult to estimate. In this study, datasets from the literature are used to model and forecast the sensitivity compressive strength of high-performance concrete. Researchers were able to examine the stability of this framework by comparing it to other models (such as Random forest, Decision tree, and Artificial neural network). For the concrete compressive strength of high-performance concrete, estimated 1030 data sets of nine input variables, such as cement, blast furnace slag, water, superplasticizer, fine aggregate, concrete age, and so on, are required. The most exciting discovery was that ANN models could be a useful tool for producing forecasts from laboratory and field data, as well as a precise and effective explicit formulation for various civil engineering applications where such assertions are required. Both of these outcomes may be related to ANN's ability to predict future events. |
Full text from SpringerLink