Optimal Design of Continuous Reinforced Concrete Beams Using Neural Networks
Journal Title: Transactions on Machine Learning and Artificial Intelligence - Year 2015, Vol 3, Issue 4
Abstract
This paper aims to build a neural network model to optimally design two-span continuous reinforced concrete beams. The training and checking data of the neural network are obtained by genetic algorithms, whose constraints are built according to the ACI Building Code and objective function is to find the minimum cost of longitudinal reinforcement, stirrups and concrete. The neural network adopted in this paper is the feed forward back propagation network, whose input vector consists of the span, width and effective depth of the beam, dead load, compressive strength of concrete as well as yield strength of steel and the output vector the positive and negative steel ratios and minimum total cost. The correlation coefficients between the target and network output of the testing data can reach as high as 0.9992, 0.9980 and 0.9999, respectively for the positive and negative steel ratios and minimum cost. Compared with the adaptive neuro-fuzzy inference system, the neural network shows almost the same accuracy but is much easier implemented.
Authors and Affiliations
Jiin-Po Yeh, Ren-Pei Yang
Automatic Tool Selection System based on STEP_NC and ISO13399 Standards
Computer Aided Process Planning (CAPP) is considered as an essential component of Computer Integrated manufacturing (CIM) environment and it is an important interface between Computer Aided Design (CAD) and Computer Aide...
Approaching Mental Disorders from the Engineering Point of View
Mental illness and mental disorders represent an increasing burden affectingthepopulation of all ages at all places, challenging mental health and health systems and contributing to the onset or to the acceleration of ma...
Electrical Behavior of Solar Cell Based on ZnO/PS
Because of high loss of photovoltaic conversion due to reflection of incident photon by the silicon surface, we proposed in this work a single heterojunction solar cell model based on porous silicon (PS) and Zinc oxide t...
Unied Acoustic Modeling using Deep Conditional Random Fields
Acoustic models based on Deep Neural Networks (DNNs) lead to sig- nicant improvement in the recognition accuracy. In these methods, Hid- den Markov Models (HMMs) state scores are computed using exible dis- criminant DN...
Comparison of Predictive Models for Transferring Stroke In-Patients to Intensive Care Unit
The cost for Intensive Care Unit (ICU) resources is extremely high and it affects healthcare budget that provides quality healthcare service for patients. Thus, the need for a predictive model for the decision to transfe...