Abstract:
The advancement of materials technology has led to production of higher grades of concrete strength. The application of High Strength Concrete “HSC” in civil engineering structures has increased significantly. The principal reasons for its popularity are economy, superior strength, increased stiffness and great durability. The production of HSC requires better quality for the basic material and additional special materials such as: silica fume, Fly Ash, super plasticizer and extra quality control procedures.
Obtaining test values (after 28-day) of the strength of concrete takes time and high cost, for these reasons Artificial Neural Networks (ANNs) were used to predict compressive strength of High Strength Concrete (HSC).
Artificial Neural Networks (ANNs) modeling technique was used in this research to predict compressive strength of High Strength Concrete “HSC”. One ANN Model was built of three layers feed-forward with back propagation system and consists of seven input nodes, seven hidden layer nodes and one output node. The ANN Model was developed for predicting compressive strength of cubes at the age 28 days by the Optimization Modeling System "Solver" in the Microsoft Office Excel 2010 and using 193 set of actual and reliable data collected from previous studies. Strength of concrete is tested after 28 days (cube test).
The studied parameters were matched with literature, and were found to be in a good agreement. Furthermore the ANN was used to study the impact of factors influencing the compressive strength.
As a result, compressive strength values of High Strength Concretes can be predicted in the multilayer feed forward artificial neural networks model without attempting any experiments in a quite short period of time with tiny error rates.
It is found that Artificial Neural Networks is a powerful tool in solving problems containing multiple variables, and has a good ability in performing parameters analysis.
