Abstract:
An Artificial Neural Network model was developed for lateral load performance of Dhajji Dewari retrofitted by inexpensive Carbon Fiber Reinforced Polymer retrofitting technique. This model was developed by considering five input variables (Strengthening, Opening, Retrofitting, Time and Load) for parametric analysis of timber walls. Developed Neural Network model is based on Layer Recurrent network type with two hidden layers and each hidden layer possess twenty nodes and one output layer having one node in the form of displacement. It can be used to compare the experimental data with the model data. Neural Network model consumes less energy and resources. It also saves time and waste materials. This developed Neural Network model is only applicable to experimental data covered by the paper followed however its exposure can be later extended on availability of data.
In second part of this research work focus was on Steel Fiber Reinforced High strength concrete (SFHSC) beams with variable confinements (CFRP and Steel sheet) to improve brittleness and flexural capacity. High strength concrete (HSC) has the added advantage over conventional concrete i, e. increased modulus of elasticity and stiffness which also increases its brittleness. To reduce its brittle behavior steel fibers are being used so that its application in construction industry may be increased. In this research work five high strength steel fiber reinforcement concrete beams with two different confinements (CFRP & Steel sheet) were made for comparison. Steel sheet and CFRP strip are also applied on beams to study improved flexural behavior.
