Prediction of concrete strength based on self-organizing fuzzy neural network
Xiaoyun Zhang, Huidong Wang, Delin Wang, Chendong Li · 2014
Compressive strength of concrete is the mostly used criterion in evaluating the performance of concrete in civil engineering. However, testing for compressive strength of concrete is complicated and time-consuming. More importantly, the test is usually performed at the 28th day. Therefore, strength prediction before the placement of concrete is highly in demand. Neural networks are introduced to predict the concrete strength, but the learning process and learning results are hard to intepret to engineers. Therefore, a new self-organizing fuzzy neural network (SOFNN) method based on clustering and extreme learning machine (ELM) optimization is proposed in this paper. A clustering-based method is used to obtain a compact network structure and the antecedent parameters of fuzzy rules automatically. ELM is used for the optimization of the consequent parameters of fuzzy rules. Simulation results show the validity and advantages of the proposed algorithm.