Adaptive neuro-fuzzy inference system on shear tensile strength of concrete modified with basalt fiber reinforced

Abstract To deal with the scarcity of research on basalt-reinforced agent-enhanced cement-based composite DTR, the development of the method of DTR forecast assessment becomes necessary. The paper's research evaluated 2 approaches: Tasmanian devil optimization (TDO) ANFIS-T and Artificial Rabbit Optimization (ARO) ANFIS-A. Its adjustment agents are highly relied on in the performance of this simulation. TDO and ARO will be interfaced with the Adaptive Neuro-Fuzzy Inference System to define the optimal combination of adjustment factors. In this regard, a reasonable number of 267 individuals from the available research were randomly selected and included in data collection that covers the three stages: training, validation, and testing. The following research represents the performance of various machine learning techniques that may be utilized for the prediction of the values of STS and BFRC by explaining their fundamental theories. The results indicated that the ANFIS-T outperformed ANFIS-A with R2 values of 0.979, 0.9893, and 0.9626 for the train, validation, and test stages, respectively, while ANFIS-A had R2 values of 0.9736, 0.9753, and 0.9568, respectively. The scores depict ANFIS-T to be more reliable and effective versus ANFIS-A, even with the remarkably high accuracy of the latter.