Compressive strength prediction of fly ash concrete by using machine learning techniques

In this research, the machine learning techniques such as, minimax probability machine regression (MPMR), relevance vector machine (RVM), genetic programming (GP), emotional neural network (ENN) and extreme learning machine (ELM) were utilized in the event of forecasting the 28 days compressive strength of fly ash concrete. In the present examination, exploratory database enveloping appropriate information recovered from a few past investigations has been made and used to prepare and approve the abovementioned MPMR, RVM, GP, ENN and ELM models. The database consists of cement, fly ash, coarse aggregate, fine aggregate, water, and water-binder ratio as the inputs whereas compressive strength of the concrete for 28 days is the output. The capability of the described models can be assessed by distinctive statistical parameters. The results from the mentioned models have been compared and decided that the MPMR model (R = 0.992) could be occupied as a decisive and authoritative data astute approach for forecasting the compressive strength of concrete which was fusion with fly ash as the admixture, thus preserving the tedious laboratory works. The accuracy of the adopted techniques was justified by comparing the distinct statistical parameters, distribution figures, and Taylor diagrams.