Abstract
ABSTRACT Reusing industrial waste materials is vital for environmental conservation and sustainability, with concrete playing a significant role in recycling efforts and air purification. This paper proposes an efficient hybrid technique for developing LWSCC incorporating waste-expanded polystyrene (EPS). The proposed hybrid method is the joint execution of the war strategy optimisation (WSO) and tree hierarchical deep convolutional neural network (THDCNN). It is commonly referred to as the WSO-THDCNN approach. The primary aim is to enhance the hardening properties of LWSCC while minimising errors in material composition. WSO optimises the concrete mix by maximising the use of recycled EPS waste, while THDCNN predicts the material behaviour and structural characteristics. The method is implemented on the MATLAB platform, yielding superior outcomes compared to existing systems such as the Deep Neural Network (DNN) the Firefly Optimization Algorithm-Radial Basis Function Neural Network (FOA-RBFNN) and the Support Vector Machine (SVM). The accuracy of the existing methods is recorded at 0.780, 0.990 and 0.850, whereas the proposed technique achieves an impressive accuracy of 0.995. Furthermore, it displays a Root Mean Square Error (RMSE) of 3.549 and an Average Relative Error (ARE) of 5.43, demonstrating its effectiveness and reliability in improving LWSCC properties.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call