Browsing by Author "Nwa-David, Chidoebere David"

Now showing 1 - 1 of 1
  • Thumbnail Image
    ItemOpen Access
    Prediction and optimization of compressive strength of concrete containing nanosized cassava peel ash using artificial neural network
    (Federal University of Technology, Owerri, 2024-02) Nwa-David, Chidoebere David
    In this work, artificial neural network (ANN) was adopted for optimization of the compressive strength of concrete containing nano-sized cassava peel ash (NCPA) as partial replacement ofcement. Levernberg-Marquardt back propagation and sigmoid function were employed in the model formulation. A total of four hundred (400) data set were presented to the network. Two hundred and forty (240) were used for training, sixty (60) were used for validation, and another sixty (60) were used for testing the network's performance. After training the network, the output and targets have an R - value of 0.99909 which is equivalent to 1. This indicates that the data used for training the network, have a good fit. Data used for this formulation were obtained experimentally. From the laboratory study, maximum compressive strength of 18.70 N/mm2 ,22.10 N/mm2 , 24.20 N/mm2, 30.10 N/mm2, 33.30 N/mm2 and 36.90 N/mm2 was achieved at a water-cement ratio of 0.75 at 19.5% replacement for 7, 14, 28, 56, 90 and 150 days curing age respectively while the corresponding ANN modelled maximum strength were 17.92 N/mm2, 22.24 N/mm2 , 24.34 N/mm2 , 30.50 N/mm2 ,33.23 N/mm2 and 36.85 N/mm2 . The predicted values were very much close to the experimental results. However, it was deduced that the replacement of cement with NCPA must not exceed 20%, if NCPA-concrete is to be used as a structural material. Evaluating the adequacy of the network with student’s T-test at 95% confidence level, proved that the model is worthy of adoption for reliable, time-effective and accurate strength-optimization of nanosized concrete.