Modeling slump of ready mix concrete using genetic algorithms assisted training of Artificial Neural Networks
作者：
Highlights：
• A two stage hybrid ANN-GA approach is presented.
• Optimal initial weights and biases for training ANN were determined using GA.
• The optimal initial weights and biases were fined tuned using BP algorithm.
• The ANN-GA hybrid model showed improved prediction accuracy and fast convergence.
• The model can be used for predicting slump of RMC in quick time.
摘要
•A two stage hybrid ANN-GA approach is presented.•Optimal initial weights and biases for training ANN were determined using GA.•The optimal initial weights and biases were fined tuned using BP algorithm.•The ANN-GA hybrid model showed improved prediction accuracy and fast convergence.•The model can be used for predicting slump of RMC in quick time.
论文关键词：Artificial Neural Networks,Genetic algorithms,Back-propagation algorithm,Lavenberg Marquardt training algorithm,Concrete slump,Ready mix concrete
论文评审过程：Available online 6 September 2014.
论文官网地址：https://doi.org/10.1016/j.eswa.2014.08.048
